NYTimes:Blogs:: Popular classification:: clusters
bdanalytics
Date: (Sun) May 17, 2015
Introduction:
Data: Source: Training: https://kaggle2.blob.core.windows.net/competitions-data/kaggle/4347/NYTimesBlogTrain.csv New: https://kaggle2.blob.core.windows.net/competitions-data/kaggle/4347/NYTimesBlogTest.csv
Time period:
Synopsis:
Based on analysis utilizing <> techniques,
[](.png)
Potential next steps include:
- Organization:
- Categorize by chunk
- Priority criteria:
- Ease of change
- Impacts report
- Cleans innards
- Bug report
- manage.missing.data chunk:
- cleaner way to manage re-splitting of training vs. new entity
- extract.features chunk:
- Add n-grams for glb_txt_vars
- “RTextTools”, “tau”, “RWeka”, and “textcat” packages
- Convert user-specified mutate code to config specs
- Add n-grams for glb_txt_vars
- fit.models chunk:
- Prediction accuracy scatter graph:
- Add tiles (raw vs. PCA)
- Use shiny for drop-down of “important” features
Use plot.ly for interactive plots ?
- Change .fit suffix of model metrics to .mdl if it’s data independent (e.g. AIC, Adj.R.Squared - is it truly data independent ?, etc.)
- move model_type parameter to myfit_mdl before indep_vars_vctr (keep all model_* together)
- create a custom model for rpart that has minbucket as a tuning parameter
varImp for randomForest crashes in caret version:6.0.41 -> submit bug report
- Probability handling for multinomials vs. desired binomial outcome
- ROCR currently supports only evaluation of binary classification tasks (version 1.0.7)
extensions toward multiclass classification are scheduled for the next release
- Skip trControl.method=“cv” for dummy classifier ?
- Add custom model to caret for a dummy (baseline) classifier (binomial & multinomial) that generates proba/outcomes which mimics the freq distribution of glb_rsp_var values; Right now glb_dmy_glm_mdl always generates most frequent outcome in training data
glm_dmy_mdl should use the same method as glm_sel_mdl until custom dummy classifer is implemented
- fit.all.training chunk:
- myplot_prediction_classification: displays ‘x’ instead of ‘+’ when there are no prediction errors
- Compare glb_sel_mdl vs. glb_fin_mdl:
- varImp
- Prediction differences (shd be minimal ?)
- Move glb_analytics_diag_plots to mydsutils.R: (+) Easier to debug (-) Too many glb vars used
- Add print(ggplot.petrinet(glb_analytics_pn) + coord_flip()) at the end of every major chunk
- Parameterize glb_analytics_pn
- Move glb_impute_missing_data to mydsutils.R: (-) Too many glb vars used; glb_<>_df reassigned
- Replicate myfit_mdl_classification features in myfit_mdl_regression
- Do non-glm methods handle interaction terms ?
- f-score computation for classifiers should be summation across outcomes (not just the desired one ?)
- Add accuracy computation to glb_dmy_mdl in predict.data.new chunk
- Why does splitting fit.data.training.all chunk into separate chunks add an overhead of ~30 secs ? It’s not rbind b/c other chunks have lower elapsed time. Is it the number of plots ?
- Incorporate code chunks in print_sessionInfo
- Test against
- projects in github.com/bdanalytics
- lectures in jhu-datascience track
Analysis:
rm(list=ls())
set.seed(12345)
options(stringsAsFactors=FALSE)
source("~/Dropbox/datascience/R/myscript.R")
source("~/Dropbox/datascience/R/mydsutils.R")
source("~/Dropbox/datascience/R/myplot.R")
source("~/Dropbox/datascience/R/mypetrinet.R")
# Gather all package requirements here
suppressPackageStartupMessages(require(doMC))
registerDoMC(4)
#packageVersion("snow")
#require(sos); findFn("cosine", maxPages=2, sortby="MaxScore")
# Analysis control global variables
glb_trnng_url <- "https://kaggle2.blob.core.windows.net/competitions-data/kaggle/4347/NYTimesBlogTrain.csv"
glb_newdt_url <- "https://kaggle2.blob.core.windows.net/competitions-data/kaggle/4347/NYTimesBlogTest.csv"
glb_out_pfx <- "NYTBlogs_clusters_"
glb_save_envir <- FALSE # or TRUE
glb_is_separate_newent_dataset <- TRUE # or TRUE
glb_split_entity_newent_datasets <- TRUE # or FALSE
glb_split_newdata_method <- "sample" # "condition" or "sample" or "copy"
glb_split_newdata_condition <- "<col_name> <condition_operator> <value>" # or NULL
glb_split_newdata_size_ratio <- 0.3 # > 0 & < 1
glb_split_sample.seed <- 123 # or any integer
glb_drop_vars <- c(NULL) # or c("<col_name>")
#glb_max_fitent_obs <- 2238 # NULL # or any integer
glb_max_fitent_obs <- NULL # or any integer
glb_is_regression <- FALSE; glb_is_classification <- TRUE; glb_is_binomial <- TRUE
glb_rsp_var_raw <- "Popular"
# for classification, the response variable has to be a factor
glb_rsp_var <- "Popular.fctr"
# if the response factor is based on numbers e.g (0/1 vs. "A"/"B"),
# caret predict(..., type="prob") crashes
glb_map_rsp_raw_to_var <- function(raw) {
relevel(factor(ifelse(raw == 1, "Y", "N")), as.factor(c("Y", "N")), ref="N")
#as.factor(paste0("B", raw))
#as.factor(raw)
}
glb_map_rsp_raw_to_var(c(1, 1, 0, 0, 0))## [1] Y Y N N N
## Levels: N Yglb_map_rsp_var_to_raw <- function(var) {
as.numeric(var) - 1
#as.numeric(var)
#levels(var)[as.numeric(var)]
#c(" <=50K", " >50K")[as.numeric(var)]
}
glb_map_rsp_var_to_raw(glb_map_rsp_raw_to_var(c(1, 1, 0, 0, 0)))## [1] 1 1 0 0 0if ((glb_rsp_var != glb_rsp_var_raw) & is.null(glb_map_rsp_raw_to_var))
stop("glb_map_rsp_raw_to_var function expected")
glb_rsp_var_out <- paste0(glb_rsp_var, ".predict.") # model_id is appended later
# List info gathered for various columns
# <col_name>: <description>; <notes>
# NewsDesk = the New York Times desk that produced the story
# SectionName = the section the article appeared in (Opinion, Arts, Technology, etc.)
# SubsectionName = the subsection the article appeared in (Education, Small Business, Room for Debate, etc.)
# Headline = the title of the article
# Snippet = a small portion of the article text
# Abstract = a summary of the blog article, written by the New York Times
# WordCount = the number of words in the article
# created WordCount.log
# PubDate = the publication date, in the format "Year-Month-Day Hour:Minute:Second"
glb_date_vars <- c("PubDate")
# UniqueID = a unique identifier for each article
glb_id_vars <- c("UniqueID")
glb_is_textual <- TRUE # vs. glb_is_numerical ???
#Sys.setlocale("LC_ALL", "C") # For english
glb_txt_vars <- c("Headline", "Snippet", "Abstract")
glb_append_stop_words <- list() # NULL # or c("<freq_word>")
# Properties:
# numrows(glb_feats_df) << numrows(glb_fitent_df)
# Select terms that appear in at least 0.2 * O(FP/FN(glb_OOBent_df))
# numrows(glb_OOBent_df) = 1.1 * numrows(glb_newent_df)
glb_sprs_thresholds <- c(0.988, 0.970, 0.970) # Generates 29, 22, 22 terms
#glb_sprs_thresholds <- c(0.990, 0.970, 0.970) # Generates 41, 22, 22 terms
#glb_sprs_thresholds <- c(0.985, 0.970, 0.970) # Generates 16, 22, 22 terms
#glb_sprs_thresholds <- c(0.975, 0.965, 0.965) # Generates 08, 14, 14 terms
#glb_sprs_thresholds <- c(0.982, 0.980, 0.980) # Generates 10, 61, 62 terms
names(glb_sprs_thresholds) <- glb_txt_vars
# List transformed vars
glb_exclude_vars_as_features <- c(NULL) # or c("<var_name>")
if (glb_is_textual)
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
glb_txt_vars)
if (glb_rsp_var_raw != glb_rsp_var)
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
glb_rsp_var_raw)
# List feats that shd be excluded due to known causation by prediction variable
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
c(NULL)) # or c("<col_name>")
# List output vars (useful during testing in console)
# glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
# grep(glb_rsp_var_out, names(glb_trnent_df), value=TRUE))
glb_impute_na_data <- TRUE # or TRUE
glb_mice_complete.seed <- 144 # or any integer
glb_cluster <- FALSE # or TRUE
glb_models_lst <- list(); glb_models_df <- data.frame()
# rpart: .rnorm messes with the models badly
# caret creates dummy vars for factor feats which messes up the tuning
# - better to feed as.numeric(<feat>.fctr) to caret
# Regression
if (glb_is_regression)
glb_models_method_vctr <- c("lm", "glm", "rpart", "rf") else
# Classification
if (glb_is_binomial)
#glb_models_method_vctr <- c("glm", "rpart", "rf") else
glb_models_method_vctr <- c("glm", "rpart") else
glb_models_method_vctr <- c("rpart", "rf")
# Baseline prediction model feature(s)
glb_Baseline_mdl_var <- NULL # or c("<col_name>")
glb_model_metric_terms <- NULL # or matrix(c(
# 0,1,2,3,4,
# 2,0,1,2,3,
# 4,2,0,1,2,
# 6,4,2,0,1,
# 8,6,4,2,0
# ), byrow=TRUE, nrow=5)
glb_model_metric <- NULL # or "<metric_name>"
glb_model_metric_maximize <- NULL # or FALSE (TRUE is not the default for both classification & regression)
glb_model_metric_smmry <- NULL # or function(data, lev=NULL, model=NULL) {
# confusion_mtrx <- t(as.matrix(confusionMatrix(data$pred, data$obs)))
# #print(confusion_mtrx)
# #print(confusion_mtrx * glb_model_metric_terms)
# metric <- sum(confusion_mtrx * glb_model_metric_terms) / nrow(data)
# names(metric) <- glb_model_metric
# return(metric)
# }
glb_tune_models_df <-
rbind(
#data.frame(parameter="cp", min=0.00005, max=0.00005, by=0.000005),
#seq(from=0.01, to=0.01, by=0.01)
#data.frame(parameter="mtry", min=080, max=100, by=10),
#data.frame(parameter="mtry", min=08, max=10, by=1),
data.frame(parameter="dummy", min=2, max=4, by=1)
)
# or NULL
glb_n_cv_folds <- 3 # or NULL
glb_clf_proba_threshold <- NULL # 0.5
# Model selection criteria
if (glb_is_regression)
glb_model_evl_criteria <- c("min.RMSE.OOB", "max.R.sq.OOB", "max.Adj.R.sq.fit")
if (glb_is_classification) {
if (glb_is_binomial)
glb_model_evl_criteria <-
c("max.Accuracy.OOB", "max.auc.OOB", "max.Kappa.OOB", "min.aic.fit") else
glb_model_evl_criteria <- c("max.Accuracy.OOB", "max.Kappa.OOB")
}
glb_sel_mdl_id <- NULL # or "<model_id_prefix>.<model_method>"
glb_fin_mdl_id <- glb_sel_mdl_id # or "Final"
# Depict process
glb_analytics_pn <- petrinet(name="glb_analytics_pn",
trans_df=data.frame(id=1:6,
name=c("data.training.all","data.new",
"model.selected","model.final",
"data.training.all.prediction","data.new.prediction"),
x=c( -5,-5,-15,-25,-25,-35),
y=c( -5, 5, 0, 0, -5, 5)
),
places_df=data.frame(id=1:4,
name=c("bgn","fit.data.training.all","predict.data.new","end"),
x=c( -0, -20, -30, -40),
y=c( 0, 0, 0, 0),
M0=c( 3, 0, 0, 0)
),
arcs_df=data.frame(
begin=c("bgn","bgn","bgn",
"data.training.all","model.selected","fit.data.training.all",
"fit.data.training.all","model.final",
"data.new","predict.data.new",
"data.training.all.prediction","data.new.prediction"),
end =c("data.training.all","data.new","model.selected",
"fit.data.training.all","fit.data.training.all","model.final",
"data.training.all.prediction","predict.data.new",
"predict.data.new","data.new.prediction",
"end","end")
))
#print(ggplot.petrinet(glb_analytics_pn))
print(ggplot.petrinet(glb_analytics_pn) + coord_flip())## Loading required package: grid
glb_analytics_avl_objs <- NULL
glb_chunks_df <- myadd_chunk(NULL, "import.data")## label step_major step_minor bgn end elapsed
## 1 import.data 1 0 12.727 NA NAStep 1.0: import data
chunk option: eval=
glb_trnent_df <- myimport_data(url=glb_trnng_url, comment="glb_trnent_df",
force_header=TRUE)## [1] "Reading file ./data/NYTimesBlogTrain.csv..."
## [1] "dimensions of data in ./data/NYTimesBlogTrain.csv: 6,532 rows x 10 cols"
## NewsDesk SectionName SubsectionName
## 1 Business Crosswords/Games
## 2 Culture Arts
## 3 Business Business Day Dealbook
## 4 Business Business Day Dealbook
## 5 Science Health
## 6 Science Health
## Headline
## 1 More School Daze
## 2 New 96-Page Murakami Work Coming in December
## 3 Public Pension Funds Stay Mum on Corporate Expats
## 4 Boot Camp for Bankers
## 5 Of Little Help to Older Knees
## 6 A Benefit of Legal Marijuana
## Snippet
## 1 A puzzle from Ethan Cooper that reminds me that a bill is due.
## 2 The Strange Library will arrive just three and a half months after Mr. Murakamis latest novel, Colorless Tsukuru Tazaki and His Years of Pilgrimage.
## 3 Public pension funds have major stakes in American companies moving overseas to cut their tax bills. But they are saying little about the strategy, which could hurt the nations tax base.
## 4 As they struggle to find new business to bolster sluggish earnings, banks consider the nations 25 million veterans and service members ideal customers.
## 5 Middle-aged and older patients are unlikely to benefit in the long term from surgery to repair tears in the meniscus, pads of cartilage in the knee, a new review of studies has found.
## 6 A new study has found evidence that legal access to marijuana is associated with fewer opioid overdose deaths, but researchers said their findings should not be used as the basis for the wide adoption of legalized cannabis.
## Abstract
## 1 A puzzle from Ethan Cooper that reminds me that a bill is due.
## 2 The Strange Library will arrive just three and a half months after Mr. Murakamis latest novel, Colorless Tsukuru Tazaki and His Years of Pilgrimage.
## 3 Public pension funds have major stakes in American companies moving overseas to cut their tax bills. But they are saying little about the strategy, which could hurt the nations tax base.
## 4 As they struggle to find new business to bolster sluggish earnings, banks consider the nations 25 million veterans and service members ideal customers.
## 5 Middle-aged and older patients are unlikely to benefit in the long term from surgery to repair tears in the meniscus, pads of cartilage in the knee, a new review of studies has found.
## 6 A new study has found evidence that legal access to marijuana is associated with fewer opioid overdose deaths, but researchers said their findings should not be used as the basis for the wide adoption of legalized cannabis.
## WordCount PubDate Popular UniqueID
## 1 508 2014-09-01 22:00:09 1 1
## 2 285 2014-09-01 21:14:07 0 2
## 3 1211 2014-09-01 21:05:36 0 3
## 4 1405 2014-09-01 20:43:34 1 4
## 5 181 2014-09-01 18:58:51 1 5
## 6 245 2014-09-01 18:52:22 1 6
## NewsDesk SectionName SubsectionName
## 226 Styles
## 995
## 3327
## 4753 Multimedia
## 4802 Business Crosswords/Games
## 6463 TStyle
## Headline
## 226 For Tavi Gevinson, Fashion Takes a Back Seat, for Now
## 995 Reconsidering What to Call an Extremist Group
## 3327 Clinton's Diagnosis of What's Wrong With Politics
## 4753 'Off Color' and on Target About Race in America
## 4802 Daniel Finkel's Circle-Toss Game
## 6463 Entering the Void
## Snippet
## 226 Tavi Gevinson, the teenage fashion star turned Broadway actress, wont be much of a player at New York Fashion Week this season.
## 995 Editors have decided to adjust how The Times refer to an Islamic extremist group that controls territory in Syria and Iraq.
## 3327 Hillary Rodham Clinton continued to laugh off questions about her presidential aspirations on Tuesday, but she did shed some light on what she thinks is wrong in Washington.
## 4753 Off Color, a New York Times video series, looks at how artists of color are making sharp social commentary about race in America through comedy and performance.
## 4802 By math educator Daniel Finkel, a puzzle thats childs play. Can you figure it out?
## 6463 The Spanish artist Miquel Barcel closely examines the basic materials of life in response to Edward Hirsch questioning his own belief in a higher power.
## Abstract
## 226 Tavi Gevinson, the teenage fashion star turned Broadway actress, wont be much of a player at New York Fashion Week this season.
## 995 Editors have decided to adjust how The Times refer to an Islamic extremist group that controls territory in Syria and Iraq.
## 3327 Hillary Rodham Clinton continued to laugh off questions about her presidential aspirations on Tuesday, but she did shed some light on what she thinks is wrong in Washington.
## 4753 Off Color, a New York Times video series, looks at how artists of color are making sharp social commentary about race in America through comedy and performance.
## 4802 By math educator Daniel Finkel, a puzzle thats childs play. Can you figure it out?
## 6463 The Spanish artist Miquel Barcel closely examines the basic materials of life in response to Edward Hirsch questioning his own belief in a higher power.
## WordCount PubDate Popular UniqueID
## 226 459 2014-09-04 16:55:57 0 226
## 995 301 2014-09-15 16:05:13 0 995
## 3327 236 2014-10-14 14:45:51 0 3327
## 4753 393 2014-11-02 05:00:13 0 4753
## 4802 1628 2014-11-03 12:00:04 1 4802
## 6463 264 2014-11-27 12:00:09 0 6463
## NewsDesk SectionName SubsectionName
## 6527 Foreign
## 6528 Opinion Room For Debate
## 6529 Foreign
## 6530 TStyle
## 6531 Multimedia
## 6532 Business
## Headline
## 6527 1914: Russians Dominate in East Poland
## 6528 Finding a Secretary of Defense
## 6529 1889: Metropolitan Opera House Reopens in New York
## 6530 The Daily Gift: Picasso Plates for Creative Dining
## 6531 Racing From New York to Barcelona
## 6532 Math Anxiety: Why Hollywood Makes Robots of Alan Turing and Other Geniuses
## Snippet
## 6527 From the International Herald Tribune archives: Russians dominate in East Poland in 1914.
## 6528 If Chuck Hagel isn't the right Pentagon chief to respond to an onslaught of global crises, who is?
## 6529 From the International Herald Tribune archives: The Metropolitan Opera House reopens in New York in 1889.
## 6530 Each day until Christmas, the editors of T share a new holiday gift idea.
## 6531 A sailboat race from New York to Barcelona was the setting for a thrilling and sometimes terrifying video about this challenging sport.
## 6532 The visionary who stares at formulas written on walls or mirrors or better yet, thin air has become a Hollywood trope. So has the depiction of the genius who cant connect with real people.
## Abstract
## 6527 From the International Herald Tribune archives: Russians dominate in East Poland in 1914.
## 6528 If Chuck Hagel isn't the right Pentagon chief to respond to an onslaught of global crises, who is?
## 6529 From the International Herald Tribune archives: The Metropolitan Opera House reopens in New York in 1889.
## 6530 Each day until Christmas, the editors of T share a new holiday gift idea.
## 6531 A sailboat race from New York to Barcelona was the setting for a thrilling and sometimes terrifying video about this challenging sport.
## 6532 The visionary who stares at formulas written on walls or mirrors or better yet, thin air has become a Hollywood trope. So has the depiction of the genius who cant connect with real people.
## WordCount PubDate Popular UniqueID
## 6527 176 2014-11-30 13:48:40 0 6527
## 6528 1597 2014-11-30 13:27:23 0 6528
## 6529 214 2014-11-30 09:44:57 0 6529
## 6530 61 2014-11-30 09:00:43 0 6530
## 6531 441 2014-11-30 09:00:22 0 6531
## 6532 921 2014-11-30 07:00:40 0 6532
## 'data.frame': 6532 obs. of 10 variables:
## $ NewsDesk : chr "Business" "Culture" "Business" "Business" ...
## $ SectionName : chr "Crosswords/Games" "Arts" "Business Day" "Business Day" ...
## $ SubsectionName: chr "" "" "Dealbook" "Dealbook" ...
## $ Headline : chr "More School Daze" "New 96-Page Murakami Work Coming in December" "Public Pension Funds Stay Mum on Corporate Expats" "Boot Camp for Bankers" ...
## $ Snippet : chr "A puzzle from Ethan Cooper that reminds me that a bill is due." "The Strange Library will arrive just three and a half months after Mr. Murakamis latest novel, Colorless Tsukuru Tazaki and His"| __truncated__ "Public pension funds have major stakes in American companies moving overseas to cut their tax bills. But they are saying little"| __truncated__ "As they struggle to find new business to bolster sluggish earnings, banks consider the nations 25 million veterans and service "| __truncated__ ...
## $ Abstract : chr "A puzzle from Ethan Cooper that reminds me that a bill is due." "The Strange Library will arrive just three and a half months after Mr. Murakamis latest novel, Colorless Tsukuru Tazaki and His"| __truncated__ "Public pension funds have major stakes in American companies moving overseas to cut their tax bills. But they are saying little"| __truncated__ "As they struggle to find new business to bolster sluggish earnings, banks consider the nations 25 million veterans and service "| __truncated__ ...
## $ WordCount : int 508 285 1211 1405 181 245 258 893 1077 188 ...
## $ PubDate : chr "2014-09-01 22:00:09" "2014-09-01 21:14:07" "2014-09-01 21:05:36" "2014-09-01 20:43:34" ...
## $ Popular : int 1 0 0 1 1 1 0 1 1 0 ...
## $ UniqueID : int 1 2 3 4 5 6 7 8 9 10 ...
## - attr(*, "comment")= chr "glb_trnent_df"
## NULLif (glb_is_separate_newent_dataset) {
glb_newent_df <- myimport_data(url=glb_newdt_url, comment="glb_newent_df",
force_header=TRUE)
# To make plots / stats / checks easier in chunk:inspectORexplore.data
glb_entity_df <- myrbind_df(glb_trnent_df, glb_newent_df);
comment(glb_entity_df) <- "glb_entity_df"
} else {
glb_entity_df <- glb_trnent_df; comment(glb_entity_df) <- "glb_entity_df"
if (!glb_split_entity_newent_datasets) {
stop("Not implemented yet")
glb_newent_df <- glb_trnent_df[sample(1:nrow(glb_trnent_df),
max(2, nrow(glb_trnent_df) / 1000)),]
} else if (glb_split_newdata_method == "condition") {
glb_newent_df <- do.call("subset",
list(glb_trnent_df, parse(text=glb_split_newdata_condition)))
glb_trnent_df <- do.call("subset",
list(glb_trnent_df, parse(text=paste0("!(",
glb_split_newdata_condition,
")"))))
} else if (glb_split_newdata_method == "sample") {
require(caTools)
set.seed(glb_split_sample.seed)
split <- sample.split(glb_trnent_df[, glb_rsp_var_raw],
SplitRatio=(1-glb_split_newdata_size_ratio))
glb_newent_df <- glb_trnent_df[!split, ]
glb_trnent_df <- glb_trnent_df[split ,]
} else if (glb_split_newdata_method == "copy") {
glb_trnent_df <- glb_entity_df
comment(glb_trnent_df) <- "glb_trnent_df"
glb_newent_df <- glb_entity_df
comment(glb_newent_df) <- "glb_newent_df"
} else stop("glb_split_newdata_method should be %in% c('condition', 'sample', 'copy')")
comment(glb_newent_df) <- "glb_newent_df"
myprint_df(glb_newent_df)
str(glb_newent_df)
if (glb_split_entity_newent_datasets) {
myprint_df(glb_trnent_df)
str(glb_trnent_df)
}
} ## [1] "Reading file ./data/NYTimesBlogTest.csv..."
## [1] "dimensions of data in ./data/NYTimesBlogTest.csv: 1,870 rows x 9 cols"
## NewsDesk SectionName SubsectionName
## 1 Culture
## 2 Culture Arts
## 3 Business Crosswords/Games
## 4 Business Business Day Dealbook
## 5 Science Health
## 6 Science Health
## Headline
## 1 'Birdman' Tops the Gothams
## 2 'Sleepy Hollow' Recap: A Not-So-Shocking Death
## 3 Drinking Buddy For Falstaff
## 4 Encouraging Public Service, Through Wall Street's 'Revolving Door'
## 5 Therapy Prevents Repeat Suicide Attempts
## 6 Hoping for a Good Death
## Snippet
## 1 The backstage tale won two awards; Citizenfour, the Edward Snowden documentary, was also a winner.
## 2 In the fall season finale, a question of where the series has many places to go.
## 3 In which Timothy Polin reveals his potty mouth.
## 4 The debate about pay for Wall Street executives who take government jobs appears to be based more on a populist shakedown than on good public policy.
## 5 Short-term psychotherapy may be an effective way to prevent repeated suicide attempts.
## 6 What I hadnt considered before my fathers heart attack was the precise meaning of not wanting to live hooked up to machines.
## Abstract
## 1 The backstage tale won two awards; Citizenfour, the Edward Snowden documentary, was also a winner.
## 2 In the fall season finale, a question of where the series has many places to go.
## 3 In which Timothy Polin reveals his potty mouth.
## 4 The debate about pay for Wall Street executives who take government jobs appears to be based more on a populist shakedown than on good public policy.
## 5 Short-term psychotherapy may be an effective way to prevent repeated suicide attempts.
## 6 What I hadnt considered before my fathers heart attack was the precise meaning of not wanting to live hooked up to machines.
## WordCount PubDate UniqueID
## 1 111 2014-12-01 22:45:24 6533
## 2 558 2014-12-01 22:01:34 6534
## 3 788 2014-12-01 22:00:26 6535
## 4 915 2014-12-01 21:04:13 6536
## 5 213 2014-12-01 19:13:20 6537
## 6 938 2014-12-01 19:05:12 6538
## NewsDesk SectionName SubsectionName
## 3 Business Crosswords/Games
## 334 OpEd Opinion
## 725 TStyle
## 732 Business Business Day Dealbook
## 752 Business Business Day Dealbook
## 864
## Headline
## 3 Drinking Buddy For Falstaff
## 334 Facts & Figures: America’s Unique Take on Maternity Leave
## 725 Ansel Elgort Buttons Up in Brioni
## 732 A Shake-Up as the Financial World Infiltrates Philanthropy
## 752 Coupang, a South Korean E-Commerce Site, Raises $300 Million
## 864 Today in Politics
## Snippet
## 3 In which Timothy Polin reveals his potty mouth.
## 334 In the U.S., paid parental leave is more of a perk than a guarantee.
## 725 The actor brought a tinge of youthfulness to the classic Italian houses retro-tailored look.
## 732 Donor-advised funds help investors get deductions for charitable donations in one year, but society doesnt get the benefit of the money right away.
## 752 The latest financing round underscores Coupangs maturity and its ambitions to one day be a publicly traded company.
## 864 The 113th Congress is concluding with partisan brinksmanship and one last mad scramble for votes to pass a $1.1 trillion spending package.
## Abstract
## 3 In which Timothy Polin reveals his potty mouth.
## 334 In the U.S., paid parental leave is more of a perk than a guarantee.
## 725 The actor brought a tinge of youthfulness to the classic Italian houses retro-tailored look.
## 732 Donor-advised funds help investors get deductions for charitable donations in one year, but society doesnt get the benefit of the money right away.
## 752 The latest financing round underscores Coupangs maturity and its ambitions to one day be a publicly traded company.
## 864 The 113th Congress is concluding with partisan brinksmanship and one last mad scramble for votes to pass a $1.1 trillion spending package.
## WordCount PubDate UniqueID
## 3 788 2014-12-01 22:00:26 6535
## 334 160 2014-12-04 11:45:20 6866
## 725 89 2014-12-10 12:30:47 7257
## 732 1172 2014-12-10 12:00:38 7264
## 752 353 2014-12-10 08:30:41 7284
## 864 1544 2014-12-11 07:09:25 7396
## NewsDesk SectionName SubsectionName
## 1865
## 1866 Business Technology
## 1867 Metro N.Y. / Region
## 1868 Multimedia
## 1869 Foreign World Asia Pacific
## 1870 Science Health
## Headline
## 1865 Today in Politics
## 1866 Uber Suspends Operations in Spain
## 1867 New York Today: The Year in News
## 1868 New Year, Old Memories, in Times Square
## 1869 Hong Kong Police Criticized After 14-Year-Old's Detention
## 1870 The Super-Short Workout and Other Fitness Trends
## Snippet
## 1865 House Republicans are ending the year on a defensive note over Representative Steve Scalises 2002 speech to a white supremacist group.
## 1866 In a first in the growing pushback against Ubers global expansion, a judges ruling barred telecommunications operators and banks from supporting the companys services.
## 1867 Wednesday: The most read stories of 2014, teeth-chattering cold, and its New Years Eve.
## 1868 What happens when you combine Burning Man, Independence Day fireworks, the last day of school and a full-contact Black Friday sale-a-bration? New Years Eve in Times Square.
## 1869 The authorities have been accused of trying to intimidate young pro-democracy protesters and their families after a 14-year-old girl was detained on suspicion of drawing flowers in chalk near government headquarters and sent to a juvenile home.
## 1870 The big story in exercise science this year was the super-short workout, although many other fitness-related themes emerged in 2014.
## Abstract
## 1865 House Republicans are ending the year on a defensive note over Representative Steve Scalises 2002 speech to a white supremacist group.
## 1866 In a first in the growing pushback against Ubers global expansion, a judges ruling barred telecommunications operators and banks from supporting the companys services.
## 1867 Wednesday: The most read stories of 2014, teeth-chattering cold, and its New Years Eve.
## 1868 What happens when you combine Burning Man, Independence Day fireworks, the last day of school and a full-contact Black Friday sale-a-bration? New Years Eve in Times Square.
## 1869 The authorities have been accused of trying to intimidate young pro-democracy protesters and their families after a 14-year-old girl was detained on suspicion of drawing flowers in chalk near government headquarters and sent to a juvenile home.
## 1870 The big story in exercise science this year was the super-short workout, although many other fitness-related themes emerged in 2014.
## WordCount PubDate UniqueID
## 1865 1616 2014-12-31 07:03:46 8397
## 1866 292 2014-12-31 06:09:32 8398
## 1867 1010 2014-12-31 06:06:58 8399
## 1868 387 2014-12-31 05:00:19 8400
## 1869 717 2014-12-31 04:16:29 8401
## 1870 818 2014-12-31 00:01:10 8402
## 'data.frame': 1870 obs. of 9 variables:
## $ NewsDesk : chr "Culture" "Culture" "Business" "Business" ...
## $ SectionName : chr "" "Arts" "Crosswords/Games" "Business Day" ...
## $ SubsectionName: chr "" "" "" "Dealbook" ...
## $ Headline : chr "'Birdman' Tops the Gothams" "'Sleepy Hollow' Recap: A Not-So-Shocking Death" "Drinking Buddy For Falstaff" "Encouraging Public Service, Through Wall Street's 'Revolving Door'" ...
## $ Snippet : chr "The backstage tale won two awards; Citizenfour, the Edward Snowden documentary, was also a winner." "In the fall season finale, a question of where the series has many places to go." "In which Timothy Polin reveals his potty mouth." "The debate about pay for Wall Street executives who take government jobs appears to be based more on a populist shakedown than "| __truncated__ ...
## $ Abstract : chr "The backstage tale won two awards; Citizenfour, the Edward Snowden documentary, was also a winner." "In the fall season finale, a question of where the series has many places to go." "In which Timothy Polin reveals his potty mouth." "The debate about pay for Wall Street executives who take government jobs appears to be based more on a populist shakedown than "| __truncated__ ...
## $ WordCount : int 111 558 788 915 213 938 1336 2644 752 99 ...
## $ PubDate : chr "2014-12-01 22:45:24" "2014-12-01 22:01:34" "2014-12-01 22:00:26" "2014-12-01 21:04:13" ...
## $ UniqueID : int 6533 6534 6535 6536 6537 6538 6539 6540 6541 6542 ...
## - attr(*, "comment")= chr "glb_newent_df"
## NULLif (nrow(glb_trnent_df) == nrow(glb_entity_df))
warning("glb_trnent_df same as glb_entity_df")
if (nrow(glb_newent_df) == nrow(glb_entity_df))
warning("glb_newent_df same as glb_entity_df")
if (length(glb_drop_vars) > 0) {
warning("dropping vars: ", paste0(glb_drop_vars, collapse=", "))
glb_entity_df <- glb_entity_df[, setdiff(names(glb_entity_df), glb_drop_vars)]
glb_trnent_df <- glb_trnent_df[, setdiff(names(glb_trnent_df), glb_drop_vars)]
glb_newent_df <- glb_newent_df[, setdiff(names(glb_newent_df), glb_drop_vars)]
}
# Check for duplicates in glb_id_vars
if (length(glb_id_vars) == 0) {
warning("using .rownames as identifiers for observations")
glb_entity_df$.rownames <- rownames(glb_entity_df)
glb_id_vars <- ".rownames"
}
if (sum(duplicated(glb_entity_df[, glb_id_vars, FALSE])) > 0)
stop(glb_id_vars, " duplicated in glb_entity_df")
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, glb_id_vars)
# Combine trnent & newent into glb_entity_df for easier manipulation
glb_trnent_df$.src <- "Train"; glb_newent_df$.src <- "Test";
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, ".src")
glb_entity_df <- myrbind_df(glb_trnent_df, glb_newent_df)
comment(glb_entity_df) <- "glb_entity_df"
glb_trnent_df <- glb_newent_df <- NULL
glb_chunks_df <- myadd_chunk(glb_chunks_df, "inspect.data", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 1 import.data 1 0 12.727 14.681 1.955
## 2 inspect.data 2 0 14.682 NA NAStep 2.0: inspect data
#print(str(glb_entity_df))
#View(glb_entity_df)
dsp_class_dstrb <- function(var) {
xtab_df <- mycreate_xtab_df(glb_entity_df, c(".src", var))
rownames(xtab_df) <- xtab_df$.src
xtab_df <- subset(xtab_df, select=-.src)
print(xtab_df)
print(xtab_df / rowSums(xtab_df, na.rm=TRUE))
}
dsp_problem_data <- function(df) {
print(sprintf("numeric data missing in %s: ",
ifelse(!is.null(df_name <- comment(df)), df_name, "")))
print(sapply(setdiff(names(df), myfind_chr_cols_df(df)),
function(col) sum(is.na(df[, col]))))
print(sprintf("numeric data w/ 0s in %s: ",
ifelse(!is.null(df_name <- comment(df)), df_name, "")))
print(sapply(setdiff(names(df), myfind_chr_cols_df(df)),
function(col) sum(df[, col] == 0, na.rm=TRUE)))
print(sprintf("numeric data w/ Infs in %s: ",
ifelse(!is.null(df_name <- comment(df)), df_name, "")))
print(sapply(setdiff(names(df), myfind_chr_cols_df(df)),
function(col) sum(df[, col] == Inf, na.rm=TRUE)))
print(sprintf("numeric data w/ NaNs in %s: ",
ifelse(!is.null(df_name <- comment(df)), df_name, "")))
print(sapply(setdiff(names(df), myfind_chr_cols_df(df)),
function(col) sum(df[, col] == NaN, na.rm=TRUE)))
print(sprintf("string data missing in %s: ",
ifelse(!is.null(df_name <- comment(df)), df_name, "")))
print(sapply(setdiff(myfind_chr_cols_df(df), ".src"),
function(col) sum(df[, col] == "")))
}
# Performed repeatedly in other chunks
glb_chk_data <- function() {
# Histogram of predictor in glb_trnent_df & glb_newent_df
print(myplot_histogram(glb_entity_df, glb_rsp_var_raw) + facet_wrap(~ .src))
if (glb_is_classification)
dsp_class_dstrb(var=ifelse(glb_rsp_var %in% names(glb_entity_df),
glb_rsp_var, glb_rsp_var_raw))
dsp_problem_data(glb_entity_df)
}
glb_chk_data()## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## Loading required package: reshape2
## Popular.0 Popular.1 Popular.NA
## Test NA NA 1870
## Train 5439 1093 NA
## Popular.0 Popular.1 Popular.NA
## Test NA NA 1
## Train 0.8326699 0.1673301 NA
## [1] "numeric data missing in glb_entity_df: "
## WordCount Popular UniqueID
## 0 1870 0
## [1] "numeric data w/ 0s in glb_entity_df: "
## WordCount Popular UniqueID
## 109 5439 0
## [1] "numeric data w/ Infs in glb_entity_df: "
## WordCount Popular UniqueID
## 0 0 0
## [1] "numeric data w/ NaNs in glb_entity_df: "
## WordCount Popular UniqueID
## 0 0 0
## [1] "string data missing in glb_entity_df: "
## NewsDesk SectionName SubsectionName Headline Snippet
## 2408 2899 6176 0 13
## Abstract PubDate
## 17 0# Create new features that help diagnostics
if (!is.null(glb_map_rsp_raw_to_var)) {
glb_entity_df[, glb_rsp_var] <-
glb_map_rsp_raw_to_var(glb_entity_df[, glb_rsp_var_raw])
mycheck_map_results(mapd_df=glb_entity_df,
from_col_name=glb_rsp_var_raw, to_col_name=glb_rsp_var)
if (glb_is_classification) dsp_class_dstrb(glb_rsp_var)
}## Loading required package: sqldf
## Loading required package: gsubfn
## Loading required package: proto
## Loading required package: RSQLite
## Loading required package: DBI
## Loading required package: tcltk## Popular Popular.fctr .n
## 1 0 N 5439
## 2 NA <NA> 1870
## 3 1 Y 1093## Warning: Removed 1 rows containing missing values (position_stack).
## Popular.fctr.N Popular.fctr.Y Popular.fctr.NA
## Test NA NA 1870
## Train 5439 1093 NA
## Popular.fctr.N Popular.fctr.Y Popular.fctr.NA
## Test NA NA 1
## Train 0.8326699 0.1673301 NA# Convert dates to numbers
# typically, dates come in as chars;
# so this must be done before converting chars to factors
myextract_dates_df <- function(df, vars, rsp_var) {
keep_feats <- c(NULL)
for (var in vars) {
dates_df <- data.frame(.date=strptime(df[, var], "%Y-%m-%d %H:%M:%S"))
dates_df[, rsp_var] <- df[, rsp_var]
dates_df[, paste0(var, ".POSIX")] <- dates_df$.date
dates_df[, paste0(var, ".year")] <- as.numeric(format(dates_df$.date, "%Y"))
dates_df[, paste0(var, ".year.fctr")] <- as.factor(format(dates_df$.date, "%Y"))
dates_df[, paste0(var, ".month")] <- as.numeric(format(dates_df$.date, "%m"))
dates_df[, paste0(var, ".month.fctr")] <- as.factor(format(dates_df$.date, "%m"))
dates_df[, paste0(var, ".date")] <- as.numeric(format(dates_df$.date, "%d"))
dates_df[, paste0(var, ".date.fctr")] <-
cut(as.numeric(format(dates_df$.date, "%d")), 5) # by month week
# wkday Sun=0; Mon=1; ...; Sat=6
dates_df[, paste0(var, ".wkday")] <- as.numeric(format(dates_df$.date, "%w"))
dates_df[, paste0(var, ".wkday.fctr")] <- as.factor(format(dates_df$.date, "%w"))
# Federal holidays 1.9., 13.10., 27.11., 25.12.
# NYState holidays 1.9., 13.10., 11.11., 27.11., 25.12.
months <- dates_df[, paste0(var, ".month")]
dates <- dates_df[, paste0(var, ".date")]
dates_df[, paste0(var, ".hlday")] <-
ifelse( ((months == 09) & (dates == 01)) |
((months == 10) & (dates == 13)) |
((months == 11) & (dates == 27)) |
((months == 12) & (dates == 25)) ,
1, 0)
dates_df[, paste0(var, ".wkend")] <- as.numeric(
(dates_df[, paste0(var, ".wkday")] %in% c(0, 6)) |
dates_df[, paste0(var, ".hlday")] )
dates_df[, paste0(var, ".hour")] <- as.numeric(format(dates_df$.date, "%H"))
dates_df[, paste0(var, ".hour.fctr")] <-
cut(as.numeric(format(dates_df$.date, "%H")), 3) # by work-shift
dates_df[, paste0(var, ".minute")] <- as.numeric(format(dates_df$.date, "%M"))
dates_df[, paste0(var, ".minute.fctr")] <-
cut(as.numeric(format(dates_df$.date, "%M")), 4) # by quarter-hours
dates_df[, paste0(var, ".second")] <- as.numeric(format(dates_df$.date, "%S"))
dates_df[, paste0(var, ".second.fctr")] <-
cut(as.numeric(format(dates_df$.date, "%S")), 4) # by quarter-hours
print(gp <- myplot_box(df=dates_df, ycol_names="PubDate.second",
xcol_name=rsp_var))
print(gp <- myplot_bar(df=dates_df, ycol_names="PubDate.second.fctr",
xcol_name=rsp_var, colorcol_name="PubDate.second.fctr"))
keep_feats <- union(keep_feats, paste(var,
c(".POSIX", ".year.fctr", ".month.fctr", ".date.fctr", ".wkday.fctr",
".wkend", ".hour.fctr", ".minute.fctr", ".second.fctr"), sep=""))
}
#myprint_df(dates_df)
return(dates_df[, keep_feats])
}
if (!is.null(glb_date_vars)) {
glb_entity_df <- cbind(glb_entity_df,
myextract_dates_df(df=glb_entity_df, vars=glb_date_vars, rsp_var=glb_rsp_var))
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
paste(glb_date_vars, c("", ".POSIX"), sep=""))
}
## Warning in mean.default(X[[1L]], ...): argument is not numeric or logical:
## returning NA## Warning in mean.default(X[[2L]], ...): argument is not numeric or logical:
## returning NA## Warning in mean.default(X[[1L]], ...): argument is not numeric or logical:
## returning NA## Warning in mean.default(X[[2L]], ...): argument is not numeric or logical:
## returning NA
srt_entity_df <- orderBy(~PubDate.POSIX, glb_entity_df)
print(myplot_scatter(subset(srt_entity_df,
PubDate.POSIX < strptime("2014-09-02", "%Y-%m-%d")),
xcol_name="PubDate.POSIX", ycol_name=glb_rsp_var,
colorcol_name=glb_rsp_var
))
# Create features that measure the gap between previous timestamp in the data
require(zoo)## Loading required package: zoo
##
## Attaching package: 'zoo'
##
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numericpd = as.POSIXlt(srt_entity_df$PubDate)
z = zoo(as.numeric(pd))
srt_entity_df[, "PubDate.zoo"] <- z
print(head(srt_entity_df))## NewsDesk SectionName SubsectionName
## 33 Science Health
## 32 Foreign World Asia Pacific
## 31 Multimedia
## 30 Culture Arts
## 29 Business Business Day Dealbook
## 28 Magazine Magazine
## Headline
## 33 Don't Catch What Ails Your House
## 32 Ukraine Conflict Has Been a Lift for China, Scholars Say
## 31 Revisiting Life and Death in Africa
## 30 Fabio Luisi Has a New Gig
## 29 Heineken to Sell Mexican Packaging Unit to Crown Holdings
## 28 Behind the Cover Story: Emily Bazelon on Medical Abortion Through the Mail
## Snippet
## 33 It doesnt take a flood to encourage the growth of mold in a home. A moist environment will do. A runny nose, coughing and all the rest typically follow.
## 32 As the United States and the European Union have imposed sanctions on Russia over the unrest in eastern Ukraine, China has been able to stand apart and gain concrete advantages, experts on foreign policy say.
## 31 Yunghi Kim went to Somalia 20 years ago expecting to cover a famine. She found herself instead in a war zone.
## 30 The music director of the Zurich Opera and principal conductor of the Metropolitan Opera will be named principal conductor of the Danish National Symphony Orchestra.
## 29 The deal values the container unit Empaque at about $1.2 billion and would make Crown Holdings the second-largest beverage can producer in North America.
## 28 Emily Bazelon, a contributing writer for the magazine, wrote this weeks cover story about the online distribution of medical abortions. Here she discusses reporting on a group of activists working to provide medical abortions through the mail.
## Abstract
## 33 It doesnt take a flood to encourage the growth of mold in a home. A moist environment will do. A runny nose, coughing and all the rest typically follow.
## 32 As the United States and the European Union have imposed sanctions on Russia over the unrest in eastern Ukraine, China has been able to stand apart and gain concrete advantages, experts on foreign policy say.
## 31 Yunghi Kim went to Somalia 20 years ago expecting to cover a famine. She found herself instead in a war zone.
## 30 The music director of the Zurich Opera and principal conductor of the Metropolitan Opera will be named principal conductor of the Danish National Symphony Orchestra.
## 29 The deal values the container unit Empaque at about $1.2 billion and would make Crown Holdings the second-largest beverage can producer in North America.
## 28 Emily Bazelon, a contributing writer for the magazine, wrote this weeks cover story about the online distribution of medical abortions. Here she discusses reporting on a group of activists working to provide medical abortions through the mail.
## WordCount PubDate Popular UniqueID .src Popular.fctr
## 33 962 2014-09-01 00:01:32 1 33 Train Y
## 32 529 2014-09-01 02:48:41 0 32 Train N
## 31 832 2014-09-01 03:00:15 0 31 Train N
## 30 166 2014-09-01 04:00:06 0 30 Train N
## 29 442 2014-09-01 04:11:20 0 29 Train N
## 28 1190 2014-09-01 05:00:26 0 28 Train N
## PubDate.POSIX PubDate.year.fctr PubDate.month.fctr
## 33 2014-09-01 00:01:32 2014 09
## 32 2014-09-01 02:48:41 2014 09
## 31 2014-09-01 03:00:15 2014 09
## 30 2014-09-01 04:00:06 2014 09
## 29 2014-09-01 04:11:20 2014 09
## 28 2014-09-01 05:00:26 2014 09
## PubDate.date.fctr PubDate.wkday.fctr PubDate.wkend PubDate.hour.fctr
## 33 (0.97,7] 1 1 (-0.023,7.67]
## 32 (0.97,7] 1 1 (-0.023,7.67]
## 31 (0.97,7] 1 1 (-0.023,7.67]
## 30 (0.97,7] 1 1 (-0.023,7.67]
## 29 (0.97,7] 1 1 (-0.023,7.67]
## 28 (0.97,7] 1 1 (-0.023,7.67]
## PubDate.minute.fctr PubDate.second.fctr PubDate.zoo
## 33 (-0.059,14.8] (29.5,44.2] 1409544092
## 32 (44.2,59.1] (29.5,44.2] 1409554121
## 31 (-0.059,14.8] (14.8,29.5] 1409554815
## 30 (-0.059,14.8] (-0.059,14.8] 1409558406
## 29 (-0.059,14.8] (14.8,29.5] 1409559080
## 28 (-0.059,14.8] (14.8,29.5] 1409562026print(myplot_scatter(subset(srt_entity_df,
PubDate.POSIX < strptime("2014-09-02", "%Y-%m-%d")),
xcol_name="PubDate.zoo", ycol_name=glb_rsp_var,
colorcol_name=glb_rsp_var
))## Don't know how to automatically pick scale for object of type zoo. Defaulting to continuous
n = nrow(srt_entity_df)
b = zoo(, seq(n))
last1 = as.numeric(merge(z-lag(z, -1), b, all = TRUE))
srt_entity_df[, "PubDate.last1"] <- last1
srt_entity_df[is.na(srt_entity_df$PubDate.last1), "PubDate.last1"] <- 0
srt_entity_df[, "PubDate.last1.log"] <- log(1 + srt_entity_df[, "PubDate.last1"])
print(gp <- myplot_box(df=subset(srt_entity_df, PubDate.last1.log > 0),
ycol_names="PubDate.last1.log",
xcol_name=glb_rsp_var))
last10 = as.numeric(merge(z-lag(z, -10), b, all = TRUE))
srt_entity_df[, "PubDate.last10"] <- last10
srt_entity_df[is.na(srt_entity_df$PubDate.last10), "PubDate.last10"] <- 0
srt_entity_df[, "PubDate.last10.log"] <- log(1 + srt_entity_df[, "PubDate.last10"])
print(gp <- myplot_box(df=subset(srt_entity_df, PubDate.last10.log > 0),
ycol_names="PubDate.last10.log",
xcol_name=glb_rsp_var))
last100 = as.numeric(merge(z-lag(z, -100), b, all = TRUE))
srt_entity_df[, "PubDate.last100"] <- last100
srt_entity_df[is.na(srt_entity_df$PubDate.last100), "PubDate.last100"] <- 0
srt_entity_df[, "PubDate.last100.log"] <- log(1 + srt_entity_df[, "PubDate.last100"])
print(gp <- myplot_box(df=subset(srt_entity_df, PubDate.last100.log > 0),
ycol_names="PubDate.last100.log",
xcol_name=glb_rsp_var))
sav_entity_df <- glb_entity_df
glb_entity_df <- srt_entity_df
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
c("PubDate.zoo", "PubDate.last1", "PubDate.last10", "PubDate.last100"))
# all2$last3 = as.numeric(merge(z-lag(z, -3), b, all = TRUE))
# all2$last5 = as.numeric(merge(z-lag(z, -5), b, all = TRUE))
# all2$last10 = as.numeric(merge(z-lag(z, -10), b, all = TRUE))
# all2$last20 = as.numeric(merge(z-lag(z, -20), b, all = TRUE))
# all2$last50 = as.numeric(merge(z-lag(z, -50), b, all = TRUE))
#
#
# # order table
# all2 = all2[order(all2$id),]
#
# ## fill in NAs
# # count averages
# na.avg = all2 %>% group_by(weekend, hour) %>% dplyr::summarise(
# last1=mean(last1, na.rm=TRUE),
# last3=mean(last3, na.rm=TRUE),
# last5=mean(last5, na.rm=TRUE),
# last10=mean(last10, na.rm=TRUE),
# last20=mean(last20, na.rm=TRUE),
# last50=mean(last50, na.rm=TRUE)
# )
#
# # fill in averages
# na.merge = merge(all2, na.avg, by=c("weekend","hour"))
# na.merge = na.merge[order(na.merge$id),]
# for(i in c("last1", "last3", "last5", "last10", "last20", "last50")) {
# y = paste0(i, ".y")
# idx = is.na(all2[[i]])
# all2[idx,][[i]] <- na.merge[idx,][[y]]
# }
# rm(na.avg, na.merge, b, i, idx, n, pd, sec, sh, y, z)
# check distribution of all numeric data
dsp_numeric_vars_dstrb <- function(vars_lst) {
for (var in vars_lst) {
print(sprintf("var: %s", var))
gp <- myplot_box(df=glb_entity_df, ycol_names=var, xcol_name=glb_rsp_var)
if (inherits(glb_entity_df[, var], "factor"))
gp <- gp + facet_wrap(reformulate(var))
print(gp)
}
}
# dsp_numeric_vars_dstrb(setdiff(names(glb_entity_df),
# union(myfind_chr_cols_df(glb_entity_df),
# c(glb_rsp_var_raw, glb_rsp_var))))
add_new_diag_feats <- function(obs_df, ref_df=glb_entity_df) {
require(plyr)
obs_df <- mutate(obs_df,
# <col_name>.NA=is.na(<col_name>),
# <col_name>.fctr=factor(<col_name>,
# as.factor(union(obs_df$<col_name>, obs_twin_df$<col_name>))),
# <col_name>.fctr=relevel(factor(<col_name>,
# as.factor(union(obs_df$<col_name>, obs_twin_df$<col_name>))),
# "<ref_val>"),
# <col2_name>.fctr=relevel(factor(ifelse(<col1_name> == <val>, "<oth_val>", "<ref_val>")),
# as.factor(c("R", "<ref_val>")),
# ref="<ref_val>"),
# This doesn't work - use sapply instead
# <col_name>.fctr_num=grep(<col_name>, levels(<col_name>.fctr)),
#
# Date.my=as.Date(strptime(Date, "%m/%d/%y %H:%M")),
# Year=year(Date.my),
# Month=months(Date.my),
# Weekday=weekdays(Date.my)
# <col_name>.log=log(1 + <col.name>),
WordCount.log = log(1 + WordCount),
# <col_name>=<table>[as.character(<col2_name>)],
# <col_name>=as.numeric(<col2_name>),
.rnorm=rnorm(n=nrow(obs_df))
)
# If levels of a factor are different across obs_df & glb_newent_df; predict.glm fails
# Transformations not handled by mutate
# obs_df$<col_name>.fctr.num <- sapply(1:nrow(obs_df),
# function(row_ix) grep(obs_df[row_ix, "<col_name>"],
# levels(obs_df[row_ix, "<col_name>.fctr"])))
#print(summary(obs_df))
#print(sapply(names(obs_df), function(col) sum(is.na(obs_df[, col]))))
return(obs_df)
}
# Add WordCount.log since WordCount is not distributed normally
glb_entity_df <- add_new_diag_feats(glb_entity_df)## Loading required package: plyrprint("Replacing WordCount with WordCount.log in potential feature set")## [1] "Replacing WordCount with WordCount.log in potential feature set"glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, "WordCount")
# Remove PubDate.year since all entity data is from 2014
# Remove PubDate.month.fctr since all newent data is from December
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
c("PubDate.year", "PubDate.month.fctr"))
# Check distributions of newly transformed / extracted vars
# Enhancement: remove vars that were displayed ealier
dsp_numeric_vars_dstrb(setdiff(names(glb_entity_df),
union(myfind_chr_cols_df(glb_entity_df),
union(glb_rsp_var_raw,
union(glb_rsp_var, glb_exclude_vars_as_features)))))## [1] "var: PubDate.year.fctr"
## [1] "var: PubDate.date.fctr"
## [1] "var: PubDate.wkday.fctr"
## [1] "var: PubDate.wkend"
## [1] "var: PubDate.hour.fctr"
## [1] "var: PubDate.minute.fctr"
## [1] "var: PubDate.second.fctr"
## [1] "var: PubDate.last1.log"
## [1] "var: PubDate.last10.log"
## [1] "var: PubDate.last100.log"
## [1] "var: WordCount.log"
## [1] "var: .rnorm"
# Convert factors to dummy variables
# Build splines require(splines); bsBasis <- bs(training$age, df=3)
#pairs(subset(glb_trnent_df, select=-c(col_symbol)))
# Check for glb_newent_df & glb_trnent_df features range mismatches
# Other diagnostics:
# print(subset(glb_trnent_df, <col1_name> == max(glb_trnent_df$<col1_name>, na.rm=TRUE) &
# <col2_name> <= mean(glb_trnent_df$<col1_name>, na.rm=TRUE)))
# print(glb_trnent_df[which.max(glb_trnent_df$<col_name>),])
# print(<col_name>_freq_glb_trnent_df <- mycreate_tbl_df(glb_trnent_df, "<col_name>"))
# print(which.min(table(glb_trnent_df$<col_name>)))
# print(which.max(table(glb_trnent_df$<col_name>)))
# print(which.max(table(glb_trnent_df$<col1_name>, glb_trnent_df$<col2_name>)[, 2]))
# print(table(glb_trnent_df$<col1_name>, glb_trnent_df$<col2_name>))
# print(table(is.na(glb_trnent_df$<col1_name>), glb_trnent_df$<col2_name>))
# print(table(sign(glb_trnent_df$<col1_name>), glb_trnent_df$<col2_name>))
# print(mycreate_xtab_df(glb_trnent_df, <col1_name>))
# print(mycreate_xtab_df(glb_trnent_df, c(<col1_name>, <col2_name>)))
# print(<col1_name>_<col2_name>_xtab_glb_trnent_df <-
# mycreate_xtab_df(glb_trnent_df, c("<col1_name>", "<col2_name>")))
# <col1_name>_<col2_name>_xtab_glb_trnent_df[is.na(<col1_name>_<col2_name>_xtab_glb_trnent_df)] <- 0
# print(<col1_name>_<col2_name>_xtab_glb_trnent_df <-
# mutate(<col1_name>_<col2_name>_xtab_glb_trnent_df,
# <col3_name>=(<col1_name> * 1.0) / (<col1_name> + <col2_name>)))
# print(<col2_name>_min_entity_arr <-
# sort(tapply(glb_trnent_df$<col1_name>, glb_trnent_df$<col2_name>, min, na.rm=TRUE)))
# print(<col1_name>_na_by_<col2_name>_arr <-
# sort(tapply(glb_trnent_df$<col1_name>.NA, glb_trnent_df$<col2_name>, mean, na.rm=TRUE)))
# Other plots:
# print(myplot_box(df=glb_trnent_df, ycol_names="<col1_name>"))
# print(myplot_box(df=glb_trnent_df, ycol_names="<col1_name>", xcol_name="<col2_name>"))
# print(myplot_line(subset(glb_trnent_df, Symbol %in% c("KO", "PG")),
# "Date.my", "StockPrice", facet_row_colnames="Symbol") +
# geom_vline(xintercept=as.numeric(as.Date("2003-03-01"))) +
# geom_vline(xintercept=as.numeric(as.Date("1983-01-01")))
# )
# print(myplot_scatter(glb_entity_df, "<col1_name>", "<col2_name>", smooth=TRUE))
# print(myplot_scatter(glb_entity_df, "<col1_name>", "<col2_name>", colorcol_name="<Pred.fctr>") +
# geom_point(data=subset(glb_entity_df, <condition>),
# mapping=aes(x=<x_var>, y=<y_var>), color="red", shape=4, size=5))
rm(srt_entity_df)
glb_chunks_df <- myadd_chunk(glb_chunks_df, "cleanse.data", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 2 inspect.data 2 0 14.682 33.457 18.775
## 3 cleanse.data 2 1 33.457 NA NAStep 2.1: cleanse data
# Options:
# 1. Not fill missing vars
# 2. Fill missing numerics with a different algorithm
# 3. Fill missing chars with data based on clusters
dsp_problem_data(glb_entity_df)## [1] "numeric data missing in : "
## WordCount Popular UniqueID
## 0 1870 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 1870 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "numeric data w/ 0s in : "
## WordCount Popular UniqueID
## 109 5439 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 378
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 7624 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 11
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 11 10 10
## PubDate.last100 PubDate.last100.log WordCount.log
## 100 100 109
## .rnorm
## 0
## [1] "numeric data w/ Infs in : "
## WordCount Popular UniqueID
## 0 0 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "numeric data w/ NaNs in : "
## WordCount Popular UniqueID
## 0 0 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "string data missing in : "
## NewsDesk SectionName SubsectionName Headline Snippet
## 2408 2899 6176 0 13
## Abstract PubDate
## 17 0warning("Forcing ", nrow(subset(glb_entity_df, WordCount.log == 0)),
" obs with WordCount.log 0s to NA")## Warning: Forcing 109 obs with WordCount.log 0s to NAglb_entity_df[glb_entity_df$WordCount.log == 0, "WordCount.log"] <- NA
dsp_problem_data(glb_entity_df)## [1] "numeric data missing in : "
## WordCount Popular UniqueID
## 0 1870 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 1870 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 109
## .rnorm
## 0
## [1] "numeric data w/ 0s in : "
## WordCount Popular UniqueID
## 109 5439 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 378
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 7624 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 11
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 11 10 10
## PubDate.last100 PubDate.last100.log WordCount.log
## 100 100 0
## .rnorm
## 0
## [1] "numeric data w/ Infs in : "
## WordCount Popular UniqueID
## 0 0 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "numeric data w/ NaNs in : "
## WordCount Popular UniqueID
## 0 0 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "string data missing in : "
## NewsDesk SectionName SubsectionName Headline Snippet
## 2408 2899 6176 0 13
## Abstract PubDate
## 17 0dsp_catgs <- function() {
print("NewsDesk:")
print(table(glb_entity_df$NewsDesk))
print("SectionName:")
print(table(glb_entity_df$SectionName))
print("SubsectionName:")
print(table(glb_entity_df$SubsectionName))
}
sel_obs <- function(Popular=NULL,
NewsDesk=NULL, SectionName=NULL, SubsectionName=NULL,
Headline.contains=NULL, Snippet.contains=NULL, Abstract.contains=NULL,
Headline.pfx=NULL, NewsDesk.nb=NULL, clusterid=NULL) {
tmp_entity_df <- glb_entity_df
# Does not work for Popular == NAs ???
if (!is.null(Popular)) {
if (is.na(Popular))
tmp_entity_df <- tmp_entity_df[is.na(tmp_entity_df$Popular), ] else
tmp_entity_df <- tmp_entity_df[tmp_entity_df$Popular == Popular, ]
}
if (!is.null(NewsDesk))
tmp_entity_df <- tmp_entity_df[tmp_entity_df$NewsDesk == NewsDesk, ]
if (!is.null(SectionName))
tmp_entity_df <- tmp_entity_df[tmp_entity_df$SectionName == SectionName, ]
if (!is.null(SubsectionName))
tmp_entity_df <- tmp_entity_df[tmp_entity_df$SubsectionName == SubsectionName, ]
if (!is.null(Headline.contains))
tmp_entity_df <-
tmp_entity_df[grep(Headline.contains, tmp_entity_df$Headline), ]
if (!is.null(Snippet.contains))
tmp_entity_df <-
tmp_entity_df[grep(Snippet.contains, tmp_entity_df$Snippet), ]
if (!is.null(Abstract.contains))
tmp_entity_df <-
tmp_entity_df[grep(Abstract.contains, tmp_entity_df$Abstract), ]
if (!is.null(Headline.pfx)) {
if (length(grep("Headline.pfx", names(tmp_entity_df), fixed=TRUE, value=TRUE))
> 0) tmp_entity_df <-
tmp_entity_df[tmp_entity_df$Headline.pfx == Headline.pfx, ] else
warning("glb_entity_df does not contain Headline.pfx; ignoring that filter")
}
if (!is.null(NewsDesk.nb)) {
if (any(grepl("NewsDesk.nb", names(tmp_entity_df), fixed=TRUE)) > 0)
tmp_entity_df <-
tmp_entity_df[tmp_entity_df$NewsDesk.nb == NewsDesk.nb, ] else
warning("glb_entity_df does not contain NewsDesk.nb; ignoring that filter")
}
if (!is.null(clusterid)) {
if (any(grepl("clusterid", names(tmp_entity_df), fixed=TRUE)) > 0)
tmp_entity_df <-
tmp_entity_df[tmp_entity_df$clusterid == clusterid, ] else
warning("glb_entity_df does not contain clusterid; ignoring that filter")
}
return(glb_entity_df$UniqueID %in% tmp_entity_df$UniqueID)
}
dsp_obs <- function(..., cols=c(NULL), all=FALSE) {
tmp_df <- glb_entity_df[sel_obs(...),
union(c("UniqueID", "Popular", "Headline"), cols), FALSE]
if(all) { print(tmp_df) } else { myprint_df(tmp_df) }
}
#dsp_obs(Popular=1, NewsDesk="", SectionName="", Headline.contains="Boehner")
# dsp_obs(Popular=1, NewsDesk="", SectionName="")
# dsp_obs(Popular=NA, NewsDesk="", SectionName="")
dsp_tbl <- function(...) {
tmp_entity_df <- glb_entity_df[sel_obs(...), ]
tmp_tbl <- table(tmp_entity_df$NewsDesk,
tmp_entity_df$SectionName,
tmp_entity_df$SubsectionName,
tmp_entity_df$Popular, useNA="ifany")
#print(names(tmp_tbl))
#print(dimnames(tmp_tbl))
print(tmp_tbl)
}
dsp_hdlxtab <- function(str)
print(mycreate_sqlxtab_df(glb_entity_df[sel_obs(Headline.contains=str), ],
c("Headline.pfx", "Headline", glb_rsp_var)))
#dsp_hdlxtab("(1914)|(1939)")
dsp_catxtab <- function(str)
print(mycreate_sqlxtab_df(glb_entity_df[sel_obs(Headline.contains=str), ],
c("Headline.pfx", "NewsDesk", "SectionName", "SubsectionName", glb_rsp_var)))
# dsp_catxtab("1914)|(1939)")
# dsp_catxtab("19(14|39|64):")
# dsp_catxtab("19..:")
# Create myCategory <- NewsDesk#SectionName#SubsectionName
# Fix some data before merging categories
glb_entity_df[sel_obs(Headline.contains="Your Turn:", NewsDesk=""),
"NewsDesk"] <- "Styles"
glb_entity_df[sel_obs(Headline.contains="School", NewsDesk="", SectionName="U.S.",
SubsectionName=""),
"SubsectionName"] <- "Education"
glb_entity_df[sel_obs(Headline.contains="Today in Small Business:", NewsDesk="Business"),
"SectionName"] <- "Business Day"
glb_entity_df[sel_obs(Headline.contains="Today in Small Business:", NewsDesk="Business"),
"SubsectionName"] <- "Small Business"
glb_entity_df[sel_obs(Headline.contains="Readers Respond:"),
"SectionName"] <- "Opinion"
glb_entity_df[sel_obs(Headline.contains="Readers Respond:"),
"SubsectionName"] <- "Room For Debate"
# glb_entity_df[sel_obs(NewsDesk="Business", SectionName="", SubsectionName="", Popular=NA),
# "SubsectionName"] <- "Small Business"
# print(glb_entity_df[glb_entity_df$UniqueID %in% c(7973),
# c("UniqueID", "Headline", "myCategory", "NewsDesk", "SectionName", "SubsectionName")])
#
# glb_entity_df[sel_obs(NewsDesk="Business", SectionName="", SubsectionName=""),
# "SectionName"] <- "Technology"
# print(glb_entity_df[glb_entity_df$UniqueID %in% c(5076, 5736, 5924, 5911, 6532),
# c("UniqueID", "Headline", "myCategory", "NewsDesk", "SectionName", "SubsectionName")])
#
# glb_entity_df[sel_obs(SectionName="Health"),
# "NewsDesk"] <- "Science"
# glb_entity_df[sel_obs(SectionName="Travel"),
# "NewsDesk"] <- "Travel"
#
# glb_entity_df[sel_obs(SubsectionName="Fashion & Style"),
# "SectionName"] <- ""
# glb_entity_df[sel_obs(SubsectionName="Fashion & Style"),
# "SubsectionName"] <- ""
# glb_entity_df[sel_obs(NewsDesk="Styles", SectionName="", SubsectionName="", Popular=1),
# "SectionName"] <- "U.S."
# print(glb_entity_df[glb_entity_df$UniqueID %in% c(5486),
# c("UniqueID", "Headline", "myCategory", "NewsDesk", "SectionName", "SubsectionName")])
#
glb_entity_df$myCategory <- paste(glb_entity_df$NewsDesk,
glb_entity_df$SectionName,
glb_entity_df$SubsectionName,
sep="#")
dsp_obs( Headline.contains="Music:"
#,NewsDesk=""
#,SectionName=""
#,SubsectionName="Fashion & Style"
#,Popular=1 #NA
,cols= c("UniqueID", "Headline", "Popular", "myCategory",
"NewsDesk", "SectionName", "SubsectionName"),
all=TRUE)## UniqueID Popular
## 305 305 0
## 844 844 1
## 1331 1331 0
## 1974 1974 0
## 2563 2563 0
## 3091 3091 0
## 3589 3589 0
## 4631 4631 0
## 5125 5125 0
## 5630 5630 0
## 6095 6095 0
## 6513 6513 1
## 6927 6927 NA
## 7473 7473 NA
## 7931 7931 NA
## 8217 8217 NA
## Headline
## 305 Friday Night Music: Lucius Covers John Lennon
## 844 Friday Night Music: Cheryl Wheeler
## 1331 Friday Night Music: Cheryl Wheeler, Summer Fly
## 1974 Friday Night Music: Quilt
## 2563 Friday Night Music: Lucius in Asheville
## 3091 Friday Night Music: Sarah Jarosz and the Milk Carton Kids
## 3589 Friday Night Music: Lucius Covers the Kinks
## 4631 Friday Night Music: Amason
## 5125 Friday Night Music: Suzanne Vega, Jacob and the Angel
## 5630 Friday Night Music: Suzanne Vega, I Never Wear White
## 6095 Friday Night Music: Jessica Hernandez and the Deltas
## 6513 Saturday Morning Music: Stay Gold
## 6927 Friday Night Music: Lucius, Monsters
## 7473 Friday Night Music: Peter Gabriel, 1993
## 7931 Friday Night Music: The Roches, Winter Wonderland
## 8217 Friday Night Music: Sarah Jarosz and Aoife O'Donovan
## myCategory NewsDesk SectionName SubsectionName
## 305 OpEd#Opinion# OpEd Opinion
## 844 OpEd#Opinion# OpEd Opinion
## 1331 OpEd#Opinion# OpEd Opinion
## 1974 OpEd#Opinion# OpEd Opinion
## 2563 OpEd#Opinion# OpEd Opinion
## 3091 OpEd#Opinion# OpEd Opinion
## 3589 OpEd#Opinion# OpEd Opinion
## 4631 OpEd#Opinion# OpEd Opinion
## 5125 OpEd#Opinion# OpEd Opinion
## 5630 OpEd#Opinion# OpEd Opinion
## 6095 OpEd#Opinion# OpEd Opinion
## 6513 OpEd#Opinion# OpEd Opinion
## 6927 OpEd#Opinion# OpEd Opinion
## 7473 #Opinion# Opinion
## 7931 OpEd#Opinion# OpEd Opinion
## 8217 OpEd#Opinion# OpEd Opiniondsp_obs( Headline.contains="."
,NewsDesk=""
,SectionName="Opinion"
,SubsectionName=""
#,Popular=1 #NA
,cols= c("UniqueID", "Headline", "Popular", "myCategory",
"NewsDesk", "SectionName", "SubsectionName"),
all=TRUE)## UniqueID Popular
## 516 516 0
## 918 918 0
## 7473 7473 NA
## 7445 7445 NA
## 7419 7419 NA
## 7505 7505 NA
## 7509 7509 NA
## Headline
## 516 This Is Life Among the Roma, Europes Forgotten People
## 918 What Might Happen If Iran Becomes America's Covert Ally?
## 7473 Friday Night Music: Peter Gabriel, 1993
## 7445 Senate Committee Bothered to Authorize War Against Islamic State
## 7419 Joe on WNYCs Money Talking
## 7505 Rev. Dr. William Barber II on Todays Protest Movements
## 7509 Did Salaita Cross the Line of Civility?
## myCategory NewsDesk SectionName SubsectionName
## 516 #Opinion# Opinion
## 918 #Opinion# Opinion
## 7473 #Opinion# Opinion
## 7445 #Opinion# Opinion
## 7419 #Opinion# Opinion
## 7505 #Opinion# Opinion
## 7509 #Opinion# Opinion# Merge some categories
glb_entity_df$myCategory <-
plyr::revalue(glb_entity_df$myCategory, c(
"#Business Day#Dealbook" = "Business#Business Day#Dealbook",
"#Business Day#Small Business" = "Business#Business Day#Small Business",
"#Crosswords/Games#" = "Business#Crosswords/Games#",
"Business##" = "Business#Technology#",
"#Open#" = "Business#Technology#",
"#Technology#" = "Business#Technology#",
"#Arts#" = "Culture#Arts#",
"Culture##" = "Culture#Arts#",
"#World#Asia Pacific" = "Foreign#World#Asia Pacific",
"Foreign##" = "Foreign#World#",
"#N.Y. / Region#" = "Metro#N.Y. / Region#",
"#Opinion#" = "OpEd#Opinion#",
"OpEd##" = "OpEd#Opinion#",
"#Health#" = "Science#Health#",
"Science##" = "Science#Health#",
"Styles##" = "Styles##Fashion",
"Styles#Health#" = "Science#Health#",
"Styles#Style#Fashion & Style" = "Styles##Fashion",
"#Travel#" = "Travel#Travel#",
"Magazine#Magazine#" = "myOther",
"National##" = "myOther",
"National#U.S.#Politics" = "myOther",
"Sports##" = "myOther",
"Sports#Sports#" = "myOther",
"#U.S.#" = "myOther",
# "Business##Small Business" = "Business#Business Day#Small Business",
#
# "#Opinion#" = "#Opinion#Room For Debate",
"##" = "##"
# "Business##" = "Business#Business Day#Dealbook",
# "Foreign#World#" = "Foreign##",
# "#Open#" = "Other",
# "#Opinion#The Public Editor" = "OpEd#Opinion#",
# "Styles#Health#" = "Styles##",
# "Styles#Style#Fashion & Style" = "Styles##",
# "#U.S.#" = "#U.S.#Education",
))
ctgry_xtab_df <- orderBy(reformulate(c("-", ".n")),
mycreate_sqlxtab_df(glb_entity_df,
c("myCategory", "NewsDesk", "SectionName", "SubsectionName", glb_rsp_var)))
# myprint_df(ctgry_xtab_df)
# write.table(ctgry_xtab_df, paste0(glb_out_pfx, "ctgry_xtab.csv"),
# row.names=FALSE)
ctgry_cast_df <- orderBy(~ -Y -NA, dcast(ctgry_xtab_df,
myCategory + NewsDesk + SectionName + SubsectionName ~
Popular.fctr, sum, value.var=".n"))
myprint_df(ctgry_cast_df)## myCategory NewsDesk SectionName SubsectionName
## 33 OpEd#Opinion# OpEd Opinion
## 36 Science#Health# Science Health
## 1 ##
## 11 Business#Crosswords/Games# Business Crosswords/Games
## 40 Styles#U.S.# Styles U.S.
## 7 Business#Business Day#Dealbook Business Business Day Dealbook
## N Y NA
## 33 113 407 141
## 36 73 119 55
## 1 1163 110 338
## 11 19 103 38
## 40 77 100 62
## 7 864 88 291
## myCategory NewsDesk SectionName
## 35 Science#Health# Science
## 17 Culture#Arts# Culture
## 16 Culture#Arts# Arts
## 8 Business#Business Day#Small Business Business Day
## 13 Business#Technology# Technology
## 28 myOther National U.S.
## SubsectionName N Y NA
## 35 0 2 2
## 17 1 0 70
## 16 0 0 11
## 8 Small Business 1 0 4
## 13 0 0 1
## 28 Politics 2 0 0
## myCategory NewsDesk SectionName SubsectionName N Y NA
## 27 myOther National 2 0 0
## 28 myOther National U.S. Politics 2 0 0
## 29 myOther Sports 1 0 0
## 30 myOther Sports Sports 1 0 0
## 37 Science#Health# Styles Health 1 0 0
## 39 Styles##Fashion Styles Style Fashion & Style 2 0 0write.table(ctgry_cast_df, paste0(glb_out_pfx, "ctgry_cast.csv"),
row.names=FALSE)
print(ctgry_sum_tbl <- table(glb_entity_df$myCategory, glb_entity_df[, glb_rsp_var],
useNA="ifany"))##
## N Y <NA>
## ## 1163 110 338
## #Multimedia# 139 2 52
## #Opinion#Room For Debate 69 7 24
## #Opinion#The Public Editor 4 16 10
## #U.S.#Education 325 0 90
## Business#Business Day#Dealbook 864 88 304
## Business#Business Day#Small Business 135 5 42
## Business#Crosswords/Games# 20 103 42
## Business#Technology# 288 51 113
## Culture#Arts# 626 50 244
## Foreign#World# 172 0 47
## Foreign#World#Asia Pacific 200 3 56
## Metro#N.Y. / Region# 181 17 67
## myOther 38 0 3
## OpEd#Opinion# 115 408 164
## Science#Health# 74 122 57
## Styles##Fashion 118 1 15
## Styles#U.S.# 77 100 62
## Travel#Travel# 116 1 35
## TStyle## 715 9 105dsp_chisq.test <- function(...) {
sel_df <- glb_entity_df[sel_obs(...) &
!is.na(glb_entity_df$Popular), ]
sel_df$.marker <- 1
ref_df <- glb_entity_df[!is.na(glb_entity_df$Popular), ]
mrg_df <- merge(ref_df[, c(glb_id_vars, "Popular")],
sel_df[, c(glb_id_vars, ".marker")], all.x=TRUE)
mrg_df[is.na(mrg_df)] <- 0
print(mrg_tbl <- table(mrg_df$.marker, mrg_df$Popular))
print("Rows:Selected; Cols:Popular")
#print(mrg_tbl)
print(chisq.test(mrg_tbl))
}
# dsp_chisq.test(Headline.contains="[Ee]bola")
# dsp_chisq.test(Snippet.contains="[Ee]bola")
# dsp_chisq.test(Abstract.contains="[Ee]bola")
# print(mycreate_sqlxtab_df(glb_entity_df[sel_obs(Headline.contains="[Ee]bola"), ],
# c(glb_rsp_var, "NewsDesk", "SectionName", "SubsectionName")))
# print(table(glb_entity_df$NewsDesk, glb_entity_df$SectionName))
# print(table(glb_entity_df$SectionName, glb_entity_df$SubsectionName))
# print(table(glb_entity_df$NewsDesk, glb_entity_df$SectionName, glb_entity_df$SubsectionName))
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
c("NewsDesk", "SectionName", "SubsectionName"))
# Copy Headline into Snipper & Abstract if they are empty
print(glb_entity_df[nchar(glb_entity_df[, "Snippet"]) == 0, c("Headline", "Snippet")])## Headline
## 2838 First Draft Focus: Off to Raise Money for Democrats
## 3728 Verbatim: Obama as Supreme Court Justice?
## 4904 Election 2014: Live Coverage
## 4994 Election 2014: Live Coverage
## 5065 First Draft Focus: Honoring a Civil War Hero
## 5029 First Draft Focus: Perry's Day in Court
## 5160 Supreme Court to Hear New Health Law Challenge
## 5254 Verbatim: Will Rick Perry Run?
## 5472 First Draft Focus: A Red Carpet Welcome
## 7164 Does Torture Work? C.I.A.'s Claims vs. Senate Panel's Findings
## 7129 First Draft Focus: Pass a Bill
## 7368 Verbatim: The People's Priorities
## 7364 First Draft Focus: Three Wise Men
## Snippet
## 2838
## 3728
## 4904
## 4994
## 5065
## 5029
## 5160
## 5254
## 5472
## 7164
## 7129
## 7368
## 7364print(glb_entity_df[glb_entity_df$Headline == glb_entity_df$Snippet,
c("UniqueID", "Headline", "Snippet")])## [1] UniqueID Headline Snippet
## <0 rows> (or 0-length row.names)glb_entity_df[nchar(glb_entity_df[, "Snippet"]) == 0, "Snippet"] <-
glb_entity_df[nchar(glb_entity_df[, "Snippet"]) == 0, "Headline"]
print(glb_entity_df[nchar(glb_entity_df[, "Abstract"]) == 0, c("Headline", "Abstract")])## Headline
## 2838 First Draft Focus: Off to Raise Money for Democrats
## 3728 Verbatim: Obama as Supreme Court Justice?
## 4904 Election 2014: Live Coverage
## 4994 Election 2014: Live Coverage
## 5065 First Draft Focus: Honoring a Civil War Hero
## 5029 First Draft Focus: Perry's Day in Court
## 5160 Supreme Court to Hear New Health Law Challenge
## 5254 Verbatim: Will Rick Perry Run?
## 5472 First Draft Focus: A Red Carpet Welcome
## 7164 Does Torture Work? C.I.A.'s Claims vs. Senate Panel's Findings
## 7129 First Draft Focus: Pass a Bill
## 7368 Verbatim: The People's Priorities
## 7364 First Draft Focus: Three Wise Men
## 7329 Obama Works the Phones to Get Funding Deal Done
## 7315 House Democrats Vent Frustration With White House
## 7310 Funding Bill Hangs in Balance as House Votes
## 7309 Spending Bill Passes House With Democratic Support
## Abstract
## 2838
## 3728
## 4904
## 4994
## 5065
## 5029
## 5160
## 5254
## 5472
## 7164
## 7129
## 7368
## 7364
## 7329
## 7315
## 7310
## 7309print(glb_entity_df[glb_entity_df$Headline == glb_entity_df$Abstract,
c("UniqueID", "Headline", "Abstract")])## [1] UniqueID Headline Abstract
## <0 rows> (or 0-length row.names)glb_entity_df[nchar(glb_entity_df[, "Abstract"]) == 0, "Abstract"] <-
glb_entity_df[nchar(glb_entity_df[, "Abstract"]) == 0, "Headline"]
# WordCount_0_df <- subset(glb_entity_df, WordCount == 0)
# table(WordCount_0_df$Popular, WordCount_0_df$WordCount, useNA="ifany")
# myprint_df(WordCount_0_df[,
# c("UniqueID", "Popular", "WordCount", "Headline")])
glb_chunks_df <- myadd_chunk(glb_chunks_df, "manage.missing.data", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 3 cleanse.data 2 1 33.457 37.515 4.059
## 4 manage.missing.data 2 2 37.516 NA NAStep 2.2: manage missing data
# print(sapply(names(glb_trnent_df), function(col) sum(is.na(glb_trnent_df[, col]))))
# print(sapply(names(glb_newent_df), function(col) sum(is.na(glb_newent_df[, col]))))
# glb_trnent_df <- na.omit(glb_trnent_df)
# glb_newent_df <- na.omit(glb_newent_df)
# df[is.na(df)] <- 0
dsp_problem_data(glb_entity_df)## [1] "numeric data missing in : "
## WordCount Popular UniqueID
## 0 1870 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 1870 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 109
## .rnorm
## 0
## [1] "numeric data w/ 0s in : "
## WordCount Popular UniqueID
## 109 5439 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 378
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 7624 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 11
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 11 10 10
## PubDate.last100 PubDate.last100.log WordCount.log
## 100 100 0
## .rnorm
## 0
## [1] "numeric data w/ Infs in : "
## WordCount Popular UniqueID
## 0 0 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "numeric data w/ NaNs in : "
## WordCount Popular UniqueID
## 0 0 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "string data missing in : "
## NewsDesk SectionName SubsectionName Headline Snippet
## 2407 2883 6156 0 0
## Abstract PubDate myCategory
## 0 0 0# Not refactored into mydsutils.R since glb_*_df might be reassigned
glb_impute_missing_data <- function() {
require(mice)
set.seed(glb_mice_complete.seed)
inp_impent_df <- glb_entity_df[, setdiff(names(glb_entity_df),
union(glb_exclude_vars_as_features, glb_rsp_var))]
print("Summary before imputation: ")
print(summary(inp_impent_df))
out_impent_df <- complete(mice(inp_impent_df))
print(summary(out_impent_df))
return(out_impent_df[, "WordCount.log"])
}
if (glb_impute_na_data)
glb_entity_df[, "WordCount.log"] <- glb_impute_missing_data()## Loading required package: mice
## Loading required package: Rcpp
## Loading required package: lattice
## mice 2.22 2014-06-10## [1] "Summary before imputation: "
## PubDate.year.fctr PubDate.date.fctr PubDate.wkday.fctr PubDate.wkend
## 2014:8402 (0.97,7]:1981 0: 378 Min. :0.0000
## (7,13] :1757 1:1605 1st Qu.:0.0000
## (13,19] :1808 2:1559 Median :0.0000
## (19,25] :1650 3:1614 Mean :0.0926
## (25,31] :1206 4:1539 3rd Qu.:0.0000
## 5:1470 Max. :1.0000
## 6: 237
## PubDate.hour.fctr PubDate.minute.fctr PubDate.second.fctr
## (-0.023,7.67]:1610 (-0.059,14.8]:3119 (-0.059,14.8]:2134
## (7.67,15.3] :4484 (14.8,29.5] :1671 (14.8,29.5] :2063
## (15.3,23] :2308 (29.5,44.2] :1995 (29.5,44.2] :2112
## (44.2,59.1] :1617 (44.2,59.1] :2093
##
##
##
## PubDate.last1.log PubDate.last10.log PubDate.last100.log WordCount.log
## Min. : 0.000 Min. : 0.000 Min. : 0.00 Min. :0.6932
## 1st Qu.: 5.263 1st Qu.: 8.516 1st Qu.:11.37 1st Qu.:5.2679
## Median : 6.292 Median : 8.868 Median :11.43 Median :5.9480
## Mean : 6.094 Mean : 9.048 Mean :11.49 Mean :5.8263
## 3rd Qu.: 7.126 3rd Qu.: 9.424 3rd Qu.:11.78 3rd Qu.:6.6067
## Max. :10.875 Max. :11.744 Max. :12.95 Max. :9.2977
## NA's :109
## .rnorm myCategory
## Min. :-3.281785 Length:8402
## 1st Qu.:-0.681275 Class :character
## Median : 0.007735 Mode :character
## Mean :-0.000264
## 3rd Qu.: 0.673409
## Max. : 3.987726
##
##
## iter imp variable
## 1 1 WordCount.log
## 1 2 WordCount.log
## 1 3 WordCount.log
## 1 4 WordCount.log
## 1 5 WordCount.log
## 2 1 WordCount.log
## 2 2 WordCount.log
## 2 3 WordCount.log
## 2 4 WordCount.log
## 2 5 WordCount.log
## 3 1 WordCount.log
## 3 2 WordCount.log
## 3 3 WordCount.log
## 3 4 WordCount.log
## 3 5 WordCount.log
## 4 1 WordCount.log
## 4 2 WordCount.log
## 4 3 WordCount.log
## 4 4 WordCount.log
## 4 5 WordCount.log
## 5 1 WordCount.log
## 5 2 WordCount.log
## 5 3 WordCount.log
## 5 4 WordCount.log
## 5 5 WordCount.log
## PubDate.year.fctr PubDate.date.fctr PubDate.wkday.fctr PubDate.wkend
## 2014:8402 (0.97,7]:1981 0: 378 Min. :0.0000
## (7,13] :1757 1:1605 1st Qu.:0.0000
## (13,19] :1808 2:1559 Median :0.0000
## (19,25] :1650 3:1614 Mean :0.0926
## (25,31] :1206 4:1539 3rd Qu.:0.0000
## 5:1470 Max. :1.0000
## 6: 237
## PubDate.hour.fctr PubDate.minute.fctr PubDate.second.fctr
## (-0.023,7.67]:1610 (-0.059,14.8]:3119 (-0.059,14.8]:2134
## (7.67,15.3] :4484 (14.8,29.5] :1671 (14.8,29.5] :2063
## (15.3,23] :2308 (29.5,44.2] :1995 (29.5,44.2] :2112
## (44.2,59.1] :1617 (44.2,59.1] :2093
##
##
##
## PubDate.last1.log PubDate.last10.log PubDate.last100.log WordCount.log
## Min. : 0.000 Min. : 0.000 Min. : 0.00 Min. :0.6931
## 1st Qu.: 5.263 1st Qu.: 8.516 1st Qu.:11.37 1st Qu.:5.2730
## Median : 6.292 Median : 8.868 Median :11.43 Median :5.9480
## Mean : 6.094 Mean : 9.048 Mean :11.49 Mean :5.8267
## 3rd Qu.: 7.126 3rd Qu.: 9.424 3rd Qu.:11.78 3rd Qu.:6.6067
## Max. :10.875 Max. :11.744 Max. :12.95 Max. :9.2977
##
## .rnorm myCategory
## Min. :-3.281785 Length:8402
## 1st Qu.:-0.681275 Class :character
## Median : 0.007735 Mode :character
## Mean :-0.000264
## 3rd Qu.: 0.673409
## Max. : 3.987726
## dsp_problem_data(glb_entity_df)## [1] "numeric data missing in : "
## WordCount Popular UniqueID
## 0 1870 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 1870 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "numeric data w/ 0s in : "
## WordCount Popular UniqueID
## 109 5439 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 378
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 7624 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 11
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 11 10 10
## PubDate.last100 PubDate.last100.log WordCount.log
## 100 100 0
## .rnorm
## 0
## [1] "numeric data w/ Infs in : "
## WordCount Popular UniqueID
## 0 0 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "numeric data w/ NaNs in : "
## WordCount Popular UniqueID
## 0 0 0
## Popular.fctr PubDate.POSIX PubDate.year.fctr
## 0 0 0
## PubDate.month.fctr PubDate.date.fctr PubDate.wkday.fctr
## 0 0 0
## PubDate.wkend PubDate.hour.fctr PubDate.minute.fctr
## 0 0 0
## PubDate.second.fctr PubDate.zoo PubDate.last1
## 0 0 0
## PubDate.last1.log PubDate.last10 PubDate.last10.log
## 0 0 0
## PubDate.last100 PubDate.last100.log WordCount.log
## 0 0 0
## .rnorm
## 0
## [1] "string data missing in : "
## NewsDesk SectionName SubsectionName Headline Snippet
## 2407 2883 6156 0 0
## Abstract PubDate myCategory
## 0 0 0glb_chunks_df <- myadd_chunk(glb_chunks_df, "encode.data", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 4 manage.missing.data 2 2 37.516 42.965 5.45
## 5 encode.data 2 3 42.966 NA NAStep 2.3: encode data
# map_<col_name>_df <- myimport_data(
# url="<map_url>",
# comment="map_<col_name>_df", print_diagn=TRUE)
# map_<col_name>_df <- read.csv(paste0(getwd(), "/data/<file_name>.csv"), strip.white=TRUE)
# glb_trnent_df <- mymap_codes(glb_trnent_df, "<from_col_name>", "<to_col_name>",
# map_<to_col_name>_df, map_join_col_name="<map_join_col_name>",
# map_tgt_col_name="<to_col_name>")
# glb_newent_df <- mymap_codes(glb_newent_df, "<from_col_name>", "<to_col_name>",
# map_<to_col_name>_df, map_join_col_name="<map_join_col_name>",
# map_tgt_col_name="<to_col_name>")
# glb_trnent_df$<col_name>.fctr <- factor(glb_trnent_df$<col_name>,
# as.factor(union(glb_trnent_df$<col_name>, glb_newent_df$<col_name>)))
# glb_newent_df$<col_name>.fctr <- factor(glb_newent_df$<col_name>,
# as.factor(union(glb_trnent_df$<col_name>, glb_newent_df$<col_name>)))
glb_chunks_df <- myadd_chunk(glb_chunks_df, "extract.features", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 5 encode.data 2 3 42.966 43.023 0.057
## 6 extract.features 3 0 43.023 NA NAStep 3.0: extract features
#```{r extract_features, cache=FALSE, eval=glb_is_textual}
# Create new features that help prediction
# <col_name>.lag.2 <- lag(zoo(glb_trnent_df$<col_name>), -2, na.pad=TRUE)
# glb_trnent_df[, "<col_name>.lag.2"] <- coredata(<col_name>.lag.2)
# <col_name>.lag.2 <- lag(zoo(glb_newent_df$<col_name>), -2, na.pad=TRUE)
# glb_newent_df[, "<col_name>.lag.2"] <- coredata(<col_name>.lag.2)
#
# glb_newent_df[1, "<col_name>.lag.2"] <- glb_trnent_df[nrow(glb_trnent_df) - 1,
# "<col_name>"]
# glb_newent_df[2, "<col_name>.lag.2"] <- glb_trnent_df[nrow(glb_trnent_df),
# "<col_name>"]
# glb_entity_df <- mutate(glb_entity_df,
# A.has.http=ifelse(grepl("http",Added,fixed=TRUE), 1, 0)
# )
#
# glb_trnent_df <- mutate(glb_trnent_df,
# )
#
# glb_newent_df <- mutate(glb_newent_df,
# )
# Create factors of string variables
print(str_vars <- myfind_chr_cols_df(glb_entity_df))## NewsDesk SectionName SubsectionName Headline
## "NewsDesk" "SectionName" "SubsectionName" "Headline"
## Snippet Abstract PubDate .src
## "Snippet" "Abstract" "PubDate" ".src"
## myCategory
## "myCategory"if (length(str_vars <- setdiff(str_vars,
glb_exclude_vars_as_features)) > 0) {
for (var in str_vars) {
warning("Creating factors of string variable: ", var,
": # of unique values: ", length(unique(glb_entity_df[, var])))
glb_entity_df[, paste0(var, ".fctr")] <- factor(glb_entity_df[, var],
as.factor(unique(glb_entity_df[, var])))
# glb_trnent_df[, paste0(var, ".fctr")] <- factor(glb_trnent_df[, var],
# as.factor(unique(glb_entity_df[, var])))
# glb_newent_df[, paste0(var, ".fctr")] <- factor(glb_newent_df[, var],
# as.factor(unique(glb_entity_df[, var])))
}
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, str_vars)
}## Warning: Creating factors of string variable: myCategory: # of unique
## values: 20if (glb_is_textual) {
require(tm)
#stop("here")
glb_corpus_lst <- list(); glb_full_DTM_lst <- list(); glb_sprs_DTM_lst <- list();
for (txt_var in glb_txt_vars) {
print(sprintf("Building corpus for %s...", txt_var))
# Combine "new york" to "newyork"
# shd be created as a tm_map::content_transformer
txt_df <- glb_entity_df[, txt_var]
txt_df <- gsub("[Nn]ew [Dd]elhi", "newdelhi", txt_df)
txt_df <- gsub("[Nn]ew [Gg]uinea", "newguinea", txt_df)
txt_df <- gsub("[Nn]ew [Jj]ersey", "newjersey", txt_df)
txt_df <- gsub("[Nn]ew [Oo]rleans", "neworleans", txt_df)
txt_df <- gsub("[Nn]ew [Yy]ear", "newyear", txt_df)
txt_df <- gsub("[Nn]ew [Yy]ork", "newyork", txt_df)
txt_df <- gsub("[Nn]ew [Zz]ealand", "newzealand", txt_df)
if (txt_var == "Headline") {
# dsp_chisq.test(Headline.contains="[Nn]ew ")
# print(head(txt_df[grep("[Nn]ew ", txt_df)]))
# print(tail(txt_df[grep("[Nn]ew ", txt_df)]))
# print(sample(txt_df[grep("[Nn]ew ", txt_df)], 5))
# print(length(txt_df[grep("[Nn]ew ", txt_df)]))
# print(txt_df[grep("[Nn]ew ", txt_df)][01:20])
# print(txt_df[grep("[Nn]ew ", txt_df)][21:40])
# print(txt_df[grep("[Nn]ew ", txt_df)][41:60])
# print(txt_df[grep("[Nn]ew ", txt_df)][61:80])
# print(txt_df[grep("[Nn]ew ", txt_df)][81:100])
# #print(length(txt_df[grep("[Nn]ew [Zz]ealand", txt_df)]))
# dsp_chisq.test(Headline.contains="[Nn]ew [Yy]ork")
# dsp_chisq.test(Headline.contains="[Re]eport")
# dsp_chisq.test(Snippet.contains="[Re]eport")
#
# dsp_chisq.test(Headline.contains="[Ww]eek")
# dsp_chisq.test(Headline.contains="[Dd]ay")
# dsp_chisq.test(Headline.contains="[Ff]ashion")
# dsp_chisq.test(Headline.contains="[Tt]oday")
# dsp_chisq.test(Headline.contains="[Dd]ail")
# dsp_chisq.test(Headline.contains="2014")
# dsp_chisq.test(Headline.contains="2015")
glb_append_stop_words[["Headline"]] <- c(NULL)
}
if (txt_var == "Snippet") {
# dsp_chisq.test(Snippet.contains="[Nn]ew ")
# print(head(txt_df[grep("[Nn]ew ", txt_df)]))
# print(tail(txt_df[grep("[Nn]ew ", txt_df)]))
# print(sample(txt_df[grep("[Nn]ew ", txt_df)], 5))
# print(length(txt_df[grep("[Nn]ew ", txt_df)]))
# print(txt_df[grep("[Nn]ew ", txt_df)][11:20])
# print(txt_df[grep("[Nn]ew ", txt_df)][21:30])
# print(txt_df[grep("[Nn]ew ", txt_df)][31:40])
# print(txt_df[grep("[Nn]ew ", txt_df)][41:50])
# print(txt_df[grep("[Nn]ew ", txt_df)][51:60])
# #print(length(txt_df[grep("[Nn]ew [Zz]ealand", txt_df)]))
# dsp_chisq.test(Snippet.contains="[Ww]ill")
# dsp_chisq.test(Snippet.contains="[Tt]ime")
# dsp_chisq.test(Snippet.contains="[Ww]eek")
# dsp_chisq.test(Snippet.contains="[Yy]ear")
# dsp_chisq.test(Snippet.contains="[Ne]w [Yy]ork")
# dsp_chisq.test(Snippet.contains="[Cc]ompan")
# dsp_chisq.test(Snippet.contains="[Oo]ne")
# dsp_chisq.test(Snippet.contains="[Rr]eport")
# dsp_chisq.test(Snippet.contains="[Pp]resid")
# dsp_chisq.test(Snippet.contains="[Ss]aid")
# dsp_chisq.test(Snippet.contains="[Cc]an")
# dsp_chisq.test(Snippet.contains="[Dd]ay")
glb_append_stop_words[["Snippet"]] <- c(NULL)
#c("can")
}
if (txt_var == "Abstract") {
# dsp_chisq.test(Abstract.contains="[Nn]ew ")
# print(head(txt_df[grep("[Nn]ew ", txt_df)]))
# print(tail(txt_df[grep("[Nn]ew ", txt_df)]))
# print(sample(txt_df[grep("[Nn]ew ", txt_df)], 5))
# print(length(txt_df[grep("[Nn]ew ", txt_df)]))
# print(txt_df[grep("[Nn]ew ", txt_df)][11:20])
# print(txt_df[grep("[Nn]ew ", txt_df)][21:30])
# print(txt_df[grep("[Nn]ew ", txt_df)][31:40])
# print(txt_df[grep("[Nn]ew ", txt_df)][41:50])
# print(txt_df[grep("[Nn]ew ", txt_df)][51:60])
# #print(length(txt_df[grep("[Nn]ew [Zz]ealand", txt_df)]))
#
# dsp_chisq.test(Abstract.contains="[Ww]ill")
# dsp_chisq.test(Abstract.contains="[Tt]ime")
# dsp_chisq.test(Abstract.contains="[Ww]eek")
# dsp_chisq.test(Abstract.contains="[Yy]ear")
# dsp_chisq.test(Abstract.contains="[Ne]w [Yy]ork")
# dsp_chisq.test(Abstract.contains="[Cc]ompan")
# dsp_chisq.test(Abstract.contains="[Oo]ne")
# dsp_chisq.test(Abstract.contains="[Rr]eport")
# dsp_chisq.test(Abstract.contains="[Pp]resid")
#
# dsp_chisq.test(Abstract.contains="[Ss]aid")
# dsp_chisq.test(Abstract.contains="[Cc]an")
# dsp_chisq.test(Abstract.contains="[Dd]ay")
# dsp_chisq.test(Abstract.contains="[Ss]tate")
# dsp_chisq.test(Abstract.contains="[Mm]ake")
# dsp_chisq.test(Abstract.contains="[Bb]ank")
glb_append_stop_words[["Abstract"]] <- c(NULL)
#c("fashion", "first", "intern", "make", "newyork", "report",
# "said", "share", "show", "state", "week", "year")
}
txt_corpus <- Corpus(VectorSource(txt_df))
txt_corpus <- tm_map(txt_corpus, tolower)
txt_corpus <- tm_map(txt_corpus, PlainTextDocument)
txt_corpus <- tm_map(txt_corpus, removePunctuation)
# txt-corpus <- tm_map(txt_corpus, content_transformer(function(x, pattern) gsub(pattern, "", x))
txt_corpus <- tm_map(txt_corpus, removeWords,
c(glb_append_stop_words[[txt_var]],
stopwords("english")))
txt_corpus <- tm_map(txt_corpus, stemDocument)
full_Tf_DTM <- DocumentTermMatrix(txt_corpus,
control=list(weighting=weightTf))
print(" Full TermMatrix:"); print(full_Tf_DTM)
full_Tf_vctr <- colSums(as.matrix(full_Tf_DTM))
names(full_Tf_vctr) <- dimnames(full_Tf_DTM)[[2]]
full_Tf_df <- as.data.frame(full_Tf_vctr)
names(full_Tf_df) <- "freq.full"
full_Tf_df$term <- rownames(full_Tf_df)
full_Tf_df <- orderBy(~ -freq.full, full_Tf_df)
sprs_Tf_DTM <- removeSparseTerms(full_Tf_DTM,
glb_sprs_thresholds[txt_var])
print(" TermMatrix:"); print(sprs_Tf_DTM)
sprs_Tf_vctr <- colSums(as.matrix(sprs_Tf_DTM))
names(sprs_Tf_vctr) <- dimnames(sprs_Tf_DTM)[[2]]
sprs_Tf_df <- as.data.frame(sprs_Tf_vctr)
names(sprs_Tf_df) <- "freq.sprs"
sprs_Tf_df$term <- rownames(sprs_Tf_df)
sprs_Tf_df <- orderBy(~ -freq.sprs, sprs_Tf_df)
terms_Tf_df <- merge(full_Tf_df, sprs_Tf_df, all.x=TRUE)
melt_Tf_df <- orderBy(~ -value, melt(terms_Tf_df, id.var="term"))
print(ggplot(melt_Tf_df, aes(value, color=variable)) + stat_ecdf() +
geom_hline(yintercept=glb_sprs_thresholds[txt_var],
linetype = "dotted"))
melt_Tf_df <- orderBy(~ -value,
melt(subset(terms_Tf_df, !is.na(freq.sprs)), id.var="term"))
print(myplot_hbar(melt_Tf_df, "term", "value",
colorcol_name="variable"))
melt_Tf_df <- orderBy(~ -value,
melt(subset(terms_Tf_df, is.na(freq.sprs)), id.var="term"))
print(myplot_hbar(head(melt_Tf_df, 10), "term", "value",
colorcol_name="variable"))
full_TfIdf_DTM <- DocumentTermMatrix(txt_corpus,
control=list(weighting=weightTfIdf))
print(" Full TermMatrix:"); print(full_TfIdf_DTM)
full_TfIdf_vctr <- colSums(as.matrix(full_TfIdf_DTM))
names(full_TfIdf_vctr) <- dimnames(full_TfIdf_DTM)[[2]]
full_TfIdf_df <- as.data.frame(full_TfIdf_vctr)
names(full_TfIdf_df) <- "freq.full"
full_TfIdf_df$term <- rownames(full_TfIdf_df)
full_TfIdf_df <- orderBy(~ -freq.full, full_TfIdf_df)
sprs_TfIdf_DTM <- removeSparseTerms(full_TfIdf_DTM,
glb_sprs_thresholds[txt_var])
print(" Sparse TermMatrix:"); print(sprs_TfIdf_DTM)
sprs_TfIdf_vctr <- colSums(as.matrix(sprs_TfIdf_DTM))
names(sprs_TfIdf_vctr) <- dimnames(sprs_TfIdf_DTM)[[2]]
sprs_TfIdf_df <- as.data.frame(sprs_TfIdf_vctr)
names(sprs_TfIdf_df) <- "freq.sprs"
sprs_TfIdf_df$term <- rownames(sprs_TfIdf_df)
sprs_TfIdf_df <- orderBy(~ -freq.sprs, sprs_TfIdf_df)
terms_TfIdf_df <- merge(full_TfIdf_df, sprs_TfIdf_df, all.x=TRUE)
melt_TfIdf_df <- orderBy(~ -value, melt(terms_TfIdf_df, id.var="term"))
print(ggplot(melt_TfIdf_df, aes(value, color=variable)) + stat_ecdf() +
geom_hline(yintercept=glb_sprs_thresholds[txt_var],
linetype = "dotted"))
melt_TfIdf_df <- orderBy(~ -value,
melt(subset(terms_TfIdf_df, !is.na(freq.sprs)), id.var="term"))
print(myplot_hbar(melt_TfIdf_df, "term", "value",
colorcol_name="variable"))
melt_TfIdf_df <- orderBy(~ -value,
melt(subset(terms_TfIdf_df, is.na(freq.sprs)), id.var="term"))
print(myplot_hbar(head(melt_TfIdf_df, 10), "term", "value",
colorcol_name="variable"))
glb_corpus_lst[[txt_var]] <- txt_corpus
# glb_full_DTM_lst[[txt_var]] <- full_Tf_DTM
# glb_sprs_DTM_lst[[txt_var]] <- sprs_Tf_DTM
glb_full_DTM_lst[[txt_var]] <- full_TfIdf_DTM
glb_sprs_DTM_lst[[txt_var]] <- sprs_TfIdf_DTM
}
# Create txt features
if ((length(glb_txt_vars) > 1) &&
(length(unique(pfxs <- sapply(glb_txt_vars,
function(txt) toupper(substr(txt, 1, 1))))) < length(glb_txt_vars)))
stop("Prefixes for corpus freq terms not unique: ", pfxs)
for (txt_var in glb_txt_vars) {
print(sprintf("Binding DTM for %s...", txt_var))
txt_var_pfx <- toupper(substr(txt_var, 1, 1))
txt_X_df <- as.data.frame(as.matrix(glb_sprs_DTM_lst[[txt_var]]))
colnames(txt_X_df) <- paste(txt_var_pfx, ".T.",
make.names(colnames(txt_X_df)), sep="")
rownames(txt_X_df) <- rownames(glb_entity_df) # warning otherwise
# plt_X_df <- cbind(txt_X_df, glb_entity_df[, c(glb_id_vars, glb_rsp_var)])
# print(myplot_box(df=plt_X_df, ycol_names="H.T.today", xcol_name=glb_rsp_var))
log_X_df <- log(1 + txt_X_df)
colnames(log_X_df) <- paste(colnames(txt_X_df), ".log", sep="")
# plt_X_df <- cbind(log_X_df, glb_entity_df[, c(glb_id_vars, glb_rsp_var)])
# print(myplot_box(df=plt_X_df, ycol_names="H.T.today.log", xcol_name=glb_rsp_var))
glb_entity_df <- cbind(glb_entity_df, txt_X_df) # TfIdf is normalized
#glb_entity_df <- cbind(glb_entity_df, log_X_df) # if using non-normalized metrics
# Create <txt_var>.has.http
glb_entity_df[, paste(txt_var_pfx, ".has.http", sep="")] <-
sapply(1:nrow(glb_entity_df),
function(row_ix) ifelse(grepl("http", glb_entity_df[row_ix, txt_var], fixed=TRUE),
1, 0))
# Create user-specified term vectors
# UniqueID == 4020, H.has.ebola
# dsp_chisq.test(Headline.contains="[Ee]bola")
# dsp_chisq.test( Snippet.contains="[Ee]bola")
# dsp_chisq.test(Abstract.contains="[Ee]bola")
if (txt_var == "Headline") {
glb_entity_df[, paste(txt_var_pfx, ".has.ebola", sep="")] <-
sapply(1:nrow(glb_entity_df),
function(row_ix) ifelse(grepl("[Ee]bola", glb_entity_df[row_ix, txt_var]),
1, 0))
}
# Create <txt_var>.nwrds.log & .nwrds.unq.log
glb_entity_df[, paste0(txt_var_pfx, ".nwrds.log")] <-
log(1 + rowSums(as.matrix(glb_full_DTM_lst[[txt_var]])))
glb_entity_df[, paste0(txt_var_pfx, ".nwrds.unq.log")] <-
log(1 + rowSums(as.matrix(glb_full_DTM_lst[[txt_var]]) != 0))
# Create <txt_var>.nchrs.log
glb_entity_df[, paste0(txt_var_pfx, ".nchrs.log")] <-
log(1 + mycount_pattern_occ(".", glb_entity_df[, txt_var]))
glb_entity_df[, paste0(txt_var_pfx, ".nuppr.log")] <-
log(1 + mycount_pattern_occ("[[:upper:]]", glb_entity_df[, txt_var]))
glb_entity_df[, paste0(txt_var_pfx, ".ndgts.log")] <-
log(1 + mycount_pattern_occ("[[:digit:]]", glb_entity_df[, txt_var]))
# Create <txt_var>.npnct?.log
punct_vctr <- c("!", "\"", "#", "\\$", "%", "&", "'", "\\(", "\\)",
"\\*", "\\+", ",", "-", "\\.", "/", ":", ";", "<", "=",
">", "\\?", "@", "\\[", "\\\\", "\\]", "^", "_", "`",
"\\{", "\\|", "\\}", "~")
for (punct_ix in 1:length(punct_vctr)) {
# smp0 <- " "
# smp1 <- "! \" # $ % & ' ( ) * + , - . / : ; < = > ? @ [ \ ] ^ _ ` { | } ~"
# smp2 <- paste(smp1, smp1, sep=" ")
# print(sprintf("Testing %s char:", punct_vctr[punct_ix]))
# results <- mycount_pattern_occ(punct_vctr[punct_ix], c(smp0, smp1, smp2))
# names(results) <- NULL
# print(results)
glb_entity_df[,
paste0(txt_var_pfx, ".npnct", sprintf("%02d", punct_ix), ".log")] <-
log(1 + mycount_pattern_occ(punct_vctr[punct_ix],
glb_entity_df[, txt_var]))
}
# print(head(glb_entity_df[glb_entity_df[, "A.npnct23.log"] > 0,
# c("UniqueID", "Popular", "Abstract", "A.npnct23.log")]))
# Create <txt_var>.has.year.colon
# mycount_pattern_occ("[0-9]{4}:", glb_entity_df$Headline[13:19])
glb_entity_df[, paste0(txt_var_pfx, ".has.year.colon")] <-
as.integer(0 + mycount_pattern_occ("[0-9]{4}:", glb_entity_df[, txt_var]))
# for (feat in paste(txt_var_pfx,
# c(".num.chars"), sep="")) {
# #print(myplot_box(glb_entity_df, paste0(feat, ".log"), glb_rsp_var))
# }
}
# Generate summaries
# print(summary(glb_entity_df))
# print(sapply(names(glb_entity_df), function(col) sum(is.na(glb_entity_df[, col]))))
# print(summary(glb_trnent_df))
# print(sapply(names(glb_trnent_df), function(col) sum(is.na(glb_trnent_df[, col]))))
# print(summary(glb_newent_df))
# print(sapply(names(glb_newent_df), function(col) sum(is.na(glb_newent_df[, col]))))
rm(full_freqs_df, melt_freqs_df, terms_freqs_df, log_X_df, txt_X_df)
}## Loading required package: tm
## Loading required package: NLP
##
## Attaching package: 'NLP'
##
## The following object is masked from 'package:ggplot2':
##
## annotate## [1] "Building corpus for Headline..."
## [1] " Full TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 9205)>>
## Non-/sparse entries: 44361/77296049
## Sparsity : 100%
## Maximal term length: 31
## Weighting : term frequency (tf)
## [1] " TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 29)>>
## Non-/sparse entries: 4686/238972
## Sparsity : 98%
## Maximal term length: 10
## Weighting : term frequency (tf)## Warning: Removed 6 rows containing missing values (geom_path).
## Warning in weighting(x): empty document(s): character(0) character(0)
## character(0)
## [1] " Full TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 9205)>>
## Non-/sparse entries: 44361/77296049
## Sparsity : 100%
## Maximal term length: 31
## Weighting : term frequency - inverse document frequency (normalized) (tf-idf)
## [1] " Sparse TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 29)>>
## Non-/sparse entries: 4686/238972
## Sparsity : 98%
## Maximal term length: 10
## Weighting : term frequency - inverse document frequency (normalized) (tf-idf)## Warning: Removed 6 rows containing missing values (geom_path).
## [1] "Building corpus for Snippet..."
## [1] " Full TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 13822)>>
## Non-/sparse entries: 105519/116026925
## Sparsity : 100%
## Maximal term length: 25
## Weighting : term frequency (tf)
## [1] " TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 22)>>
## Non-/sparse entries: 8657/176187
## Sparsity : 95%
## Maximal term length: 7
## Weighting : term frequency (tf)## Warning: Removed 6 rows containing missing values (geom_path).
## Warning in weighting(x): empty document(s): character(0) character(0)
## [1] " Full TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 13822)>>
## Non-/sparse entries: 105519/116026925
## Sparsity : 100%
## Maximal term length: 25
## Weighting : term frequency - inverse document frequency (normalized) (tf-idf)
## [1] " Sparse TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 22)>>
## Non-/sparse entries: 8657/176187
## Sparsity : 95%
## Maximal term length: 7
## Weighting : term frequency - inverse document frequency (normalized) (tf-idf)## Warning: Removed 6 rows containing missing values (geom_path).
## [1] "Building corpus for Abstract..."
## [1] " Full TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 13866)>>
## Non-/sparse entries: 105900/116396232
## Sparsity : 100%
## Maximal term length: 112
## Weighting : term frequency (tf)
## [1] " TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 22)>>
## Non-/sparse entries: 8672/176172
## Sparsity : 95%
## Maximal term length: 7
## Weighting : term frequency (tf)## Warning: Removed 6 rows containing missing values (geom_path).
## Warning in weighting(x): empty document(s): character(0) character(0)
## [1] " Full TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 13866)>>
## Non-/sparse entries: 105900/116396232
## Sparsity : 100%
## Maximal term length: 112
## Weighting : term frequency - inverse document frequency (normalized) (tf-idf)
## [1] " Sparse TermMatrix:"
## <<DocumentTermMatrix (documents: 8402, terms: 22)>>
## Non-/sparse entries: 8672/176172
## Sparsity : 95%
## Maximal term length: 7
## Weighting : term frequency - inverse document frequency (normalized) (tf-idf)## Warning: Removed 6 rows containing missing values (geom_path).
## [1] "Binding DTM for Headline..."
## [1] "Binding DTM for Snippet..."
## [1] "Binding DTM for Abstract..."## Warning in rm(full_freqs_df, melt_freqs_df, terms_freqs_df, log_X_df,
## txt_X_df): object 'full_freqs_df' not found## Warning in rm(full_freqs_df, melt_freqs_df, terms_freqs_df, log_X_df,
## txt_X_df): object 'melt_freqs_df' not found## Warning in rm(full_freqs_df, melt_freqs_df, terms_freqs_df, log_X_df,
## txt_X_df): object 'terms_freqs_df' not found
# print(sapply(names(glb_trnent_df), function(col) sum(is.na(glb_trnent_df[, col]))))
# print(sapply(names(glb_newent_df), function(col) sum(is.na(glb_newent_df[, col]))))
# print(myplot_scatter(glb_trnent_df, "<col1_name>", "<col2_name>", smooth=TRUE))
replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.training.all","data.new")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
glb_chunks_df <- myadd_chunk(glb_chunks_df, "cluster.data", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 6 extract.features 3 0 43.023 177.899 134.876
## 7 cluster.data 4 0 177.900 NA NAStep 4.0: cluster data
require(proxy)## Loading required package: proxy
##
## Attaching package: 'proxy'
##
## The following objects are masked from 'package:stats':
##
## as.dist, dist
##
## The following object is masked from 'package:base':
##
## as.matrixrequire(dynamicTreeCut)## Loading required package: dynamicTreeCutglb_entity_df$clusterid <- 1
if (glb_cluster) {
for (myCategory in c("Science#Health#")) {
ctgry_entity_df <- glb_entity_df[glb_entity_df$myCategory == myCategory, ]
cluster_vars <- grep("[HSA]\\.T\\.", names(ctgry_entity_df), value=TRUE)
dstns_dist <- dist(ctgry_entity_df[, cluster_vars], method = "cosine")
dstns_mtrx <- as.matrix(dstns_dist)
print(sprintf("max distance(%0.4f) pair:", max(dstns_mtrx)))
row_ix <- ceiling(which.max(dstns_mtrx) / ncol(dstns_mtrx))
col_ix <- which.max(dstns_mtrx[row_ix, ])
print(ctgry_entity_df[c(row_ix, col_ix),
c("UniqueID", "Popular", "myCategory", "Headline", cluster_vars)])
min_dstns_mtrx <- dstns_mtrx
diag(min_dstns_mtrx) <- 1
print(sprintf("min distance(%0.4f) pair:", min(min_dstns_mtrx)))
row_ix <- ceiling(which.min(min_dstns_mtrx) / ncol(min_dstns_mtrx))
col_ix <- which.min(min_dstns_mtrx[row_ix, ])
print(ctgry_entity_df[c(row_ix, col_ix),
c("UniqueID", "Popular", "myCategory", "Headline", cluster_vars)])
clusters <- hclust(dstns_dist, method = "ward.D2")
plot(clusters, hang=-1)
#clusterGroups = cutree(clusters, k=7)
clusterGroups <- cutreeDynamic(clusters, minClusterSize=20, method="tree", deepSplit=0)
table(clusterGroups)
clusterGroups[clusterGroups==0] = 1
table(clusterGroups)
#summary(factor(clusterGroups))
# add to glb_entity_df - then split the data again
glb_entity_df[glb_entity_df$myCategory==myCategory,]$clusterid <- clusterGroups
}
ctgry_xtab_df <- orderBy(reformulate(c("-", ".n")),
mycreate_sqlxtab_df(glb_entity_df,
c("myCategory", "clusterid", glb_rsp_var)))
ctgry_cast_df <- orderBy(~ -Y -NA, dcast(ctgry_xtab_df,
myCategory + clusterid ~
Popular.fctr, sum, value.var=".n"))
print(ctgry_cast_df)
# write.table(ctgry_cast_df, paste0(glb_out_pfx, "ctgry_clst.csv"),
# row.names=FALSE)
print(ctgry_sum_tbl <- table(glb_entity_df$myCategory, glb_entity_df$clusterid,
glb_entity_df[, glb_rsp_var],
useNA="ifany"))
dsp_obs(clusterid=18, cols=c("UniqueID", "Popular", "myCategory", "clusterid", "Headline"),
all=TRUE)
glb_entity_df$clusterid.fctr <- as.factor(glb_entity_df$clusterid)
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, "clusterid")
}
# Re-partition
glb_trnent_df <- subset(glb_entity_df, .src == "Train")
glb_newent_df <- subset(glb_entity_df, .src == "Test")
glb_chunks_df <- myadd_chunk(glb_chunks_df, "select.features", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 7 cluster.data 4 0 177.900 182.126 4.226
## 8 select.features 5 0 182.126 NA NAStep 5.0: select features
print(glb_feats_df <- myselect_features(entity_df=glb_trnent_df,
exclude_vars_as_features=glb_exclude_vars_as_features,
rsp_var=glb_rsp_var))## Warning in cor(data.matrix(entity_df[, sel_feats]), y =
## as.numeric(entity_df[, : the standard deviation is zero## id cor.y exclude.as.feat
## Popular Popular 1.000000e+00 1
## A.nuppr.log A.nuppr.log -2.720962e-01 0
## S.nuppr.log S.nuppr.log -2.718459e-01 0
## WordCount.log WordCount.log 2.656836e-01 0
## WordCount WordCount 2.575265e-01 1
## S.nwrds.unq.log S.nwrds.unq.log -2.507969e-01 0
## A.nwrds.unq.log A.nwrds.unq.log -2.506012e-01 0
## S.nchrs.log S.nchrs.log -2.246930e-01 0
## A.nchrs.log A.nchrs.log -2.245488e-01 0
## H.nwrds.unq.log H.nwrds.unq.log -2.044964e-01 0
## H.nchrs.log H.nchrs.log -1.710624e-01 0
## H.nwrds.log H.nwrds.log 1.410282e-01 0
## S.nwrds.log S.nwrds.log 1.359149e-01 0
## PubDate.hour.fctr PubDate.hour.fctr 1.354368e-01 0
## A.nwrds.log A.nwrds.log 1.354108e-01 0
## H.npnct21.log H.npnct21.log 1.283641e-01 0
## H.nuppr.log H.nuppr.log -1.278085e-01 0
## A.ndgts.log A.ndgts.log -1.249484e-01 0
## S.ndgts.log S.ndgts.log -1.242046e-01 0
## H.ndgts.log H.ndgts.log -1.196633e-01 0
## PubDate.wkend PubDate.wkend 1.067288e-01 0
## A.npnct12.log A.npnct12.log -9.183870e-02 0
## S.npnct12.log S.npnct12.log -9.158156e-02 0
## H.npnct30.log H.npnct30.log -8.917338e-02 0
## S.T.week S.T.week -8.503373e-02 0
## A.T.week A.T.week -8.492895e-02 0
## S.T.fashion S.T.fashion -8.419711e-02 0
## A.T.fashion A.T.fashion -8.419345e-02 0
## H.npnct16.log H.npnct16.log -8.273237e-02 0
## H.T.fashion H.T.fashion -7.947505e-02 0
## H.has.year.colon H.has.year.colon -7.842875e-02 0
## S.T.intern S.T.intern -6.953750e-02 0
## A.T.intern A.T.intern -6.949870e-02 0
## A.npnct16.log A.npnct16.log -6.893301e-02 0
## H.T.week H.T.week -6.812724e-02 0
## S.npnct16.log S.npnct16.log -6.770952e-02 0
## H.T.X2015 H.T.X2015 -6.601141e-02 0
## H.T.daili H.T.daili -6.299948e-02 0
## S.npnct04.log S.npnct04.log -6.294642e-02 0
## A.npnct04.log A.npnct04.log -6.294642e-02 0
## H.T.report H.T.report -6.244050e-02 0
## H.npnct15.log H.npnct15.log -6.158577e-02 0
## H.T.day H.T.day -6.033488e-02 0
## H.T.springsumm H.T.springsumm -5.943248e-02 0
## H.T.today H.T.today -5.831308e-02 0
## S.T.newyork S.T.newyork -5.712853e-02 0
## A.T.newyork A.T.newyork -5.706083e-02 0
## H.T.newyork H.T.newyork -5.650839e-02 0
## S.npnct21.log S.npnct21.log 5.503894e-02 0
## A.npnct21.log A.npnct21.log 5.482747e-02 0
## S.T.articl S.T.articl -5.446201e-02 0
## A.T.articl A.T.articl -5.445243e-02 0
## PubDate.last10 PubDate.last10 5.398093e-02 1
## H.npnct08.log H.npnct08.log 5.375262e-02 0
## H.npnct09.log H.npnct09.log 5.375262e-02 0
## S.npnct14.log S.npnct14.log -5.332519e-02 0
## A.T.time A.T.time -5.313395e-02 0
## S.T.time S.T.time -5.303654e-02 0
## H.npnct04.log H.npnct04.log -5.126277e-02 0
## S.T.share S.T.share -5.070234e-02 0
## A.T.share A.T.share -5.070234e-02 0
## A.npnct14.log A.npnct14.log -4.999563e-02 0
## PubDate.last10.log PubDate.last10.log 4.931702e-02 0
## H.T.busi H.T.busi -4.901905e-02 0
## H.T.morn H.T.morn -4.838380e-02 0
## S.T.compani S.T.compani -4.764341e-02 0
## A.T.compani A.T.compani -4.751471e-02 0
## S.T.report S.T.report -4.746920e-02 0
## A.T.report A.T.report -4.741555e-02 0
## S.T.year S.T.year -4.729011e-02 0
## A.T.year A.T.year -4.721236e-02 0
## PubDate.last1.log PubDate.last1.log 4.635751e-02 0
## S.T.first S.T.first -4.617532e-02 0
## A.T.first A.T.first -4.603341e-02 0
## H.T.X2014 H.T.X2014 -4.497745e-02 0
## H.T.first H.T.first -4.458885e-02 0
## H.T.news H.T.news -4.415284e-02 0
## A.npnct30.log A.npnct30.log -4.373349e-02 0
## S.npnct30.log S.npnct30.log -4.370037e-02 0
## H.T.new H.T.new -4.327803e-02 0
## A.T.day A.T.day -4.196599e-02 0
## A.T.show A.T.show -4.196129e-02 0
## S.T.show S.T.show -4.193803e-02 0
## S.T.day S.T.day -4.188671e-02 0
## H.T.pictur H.T.pictur -3.993172e-02 0
## PubDate.last100 PubDate.last100 3.989229e-02 1
## PubDate.wkday.fctr PubDate.wkday.fctr -3.980129e-02 0
## A.T.make A.T.make 3.965722e-02 0
## S.T.make S.T.make 3.959645e-02 0
## S.T.will S.T.will -3.892267e-02 0
## A.T.will A.T.will -3.887937e-02 0
## A.npnct13.log A.npnct13.log -3.760012e-02 0
## S.npnct13.log S.npnct13.log -3.638891e-02 0
## PubDate.last1 PubDate.last1 3.592267e-02 1
## PubDate.minute.fctr PubDate.minute.fctr -3.407385e-02 0
## H.T.art H.T.art -3.280483e-02 0
## H.npnct06.log H.npnct06.log 3.190718e-02 0
## H.T.china H.T.china -3.144808e-02 0
## A.T.can A.T.can 3.127063e-02 0
## S.npnct01.log S.npnct01.log 3.093101e-02 0
## A.npnct01.log A.npnct01.log 3.093101e-02 0
## H.T.polit H.T.polit -3.062866e-02 0
## S.T.can S.T.can 3.049697e-02 0
## H.npnct17.log H.npnct17.log 3.039622e-02 0
## H.T.billion H.T.billion -2.949817e-02 0
## A.T.new A.T.new -2.782876e-02 0
## S.T.new S.T.new -2.769558e-02 0
## S.npnct23.log S.npnct23.log 2.760321e-02 0
## S.npnct25.log S.npnct25.log 2.760321e-02 0
## H.has.ebola H.has.ebola 2.588140e-02 0
## H.T.deal H.T.deal -2.559418e-02 0
## H.npnct14.log H.npnct14.log -2.524770e-02 0
## A.npnct15.log A.npnct15.log -2.407715e-02 0
## S.npnct06.log S.npnct06.log -2.389145e-02 0
## A.npnct06.log A.npnct06.log -2.389145e-02 0
## A.T.take A.T.take -2.282555e-02 0
## S.T.take S.T.take -2.275732e-02 0
## H.npnct01.log H.npnct01.log 2.271577e-02 0
## S.npnct15.log S.npnct15.log -2.121844e-02 0
## H.T.test H.T.test -2.065255e-02 0
## H.npnct02.log H.npnct02.log -2.001851e-02 0
## S.npnct22.log S.npnct22.log -1.923169e-02 0
## A.npnct22.log A.npnct22.log -1.923169e-02 0
## PubDate.month.fctr PubDate.month.fctr 1.914874e-02 1
## A.T.said A.T.said 1.839871e-02 0
## S.T.said S.T.said 1.826884e-02 0
## S.has.year.colon S.has.year.colon -1.755336e-02 0
## A.has.year.colon A.has.year.colon -1.755336e-02 0
## PubDate.POSIX PubDate.POSIX 1.568326e-02 1
## PubDate.zoo PubDate.zoo 1.568326e-02 1
## A.npnct23.log A.npnct23.log 1.537569e-02 0
## A.npnct25.log A.npnct25.log 1.537569e-02 0
## A.npnct02.log A.npnct02.log -1.451467e-02 0
## A.npnct18.log A.npnct18.log -1.451467e-02 0
## A.npnct20.log A.npnct20.log -1.451467e-02 0
## H.T.make H.T.make 1.430572e-02 0
## H.T.big H.T.big -1.390748e-02 0
## A.has.http A.has.http -1.359260e-02 0
## A.npnct03.log A.npnct03.log -1.359260e-02 0
## H.npnct12.log H.npnct12.log 1.333613e-02 0
## H.npnct13.log H.npnct13.log -1.305305e-02 0
## A.npnct19.log A.npnct19.log -1.271661e-02 0
## S.npnct03.log S.npnct03.log -1.240734e-02 0
## myCategory.fctr myCategory.fctr 1.234541e-02 0
## S.npnct07.log S.npnct07.log -1.214357e-02 0
## A.npnct07.log A.npnct07.log -1.214357e-02 0
## H.npnct07.log H.npnct07.log -1.201741e-02 0
## PubDate.second.fctr PubDate.second.fctr -1.187946e-02 0
## UniqueID UniqueID 1.182492e-02 1
## PubDate.date.fctr PubDate.date.fctr -1.164756e-02 0
## A.T.one A.T.one 1.081694e-02 0
## S.T.one S.T.one 1.080534e-02 0
## A.T.state A.T.state 1.020706e-02 0
## S.T.state S.T.state 1.012205e-02 0
## H.T.bank H.T.bank -9.989139e-03 0
## H.T.say H.T.say -9.960773e-03 0
## H.T.obama H.T.obama -9.907543e-03 0
## H.npnct05.log H.npnct05.log -9.653967e-03 0
## H.npnct03.log H.npnct03.log 9.533020e-03 0
## .rnorm .rnorm -8.244230e-03 0
## PubDate.last100.log PubDate.last100.log -7.663322e-03 0
## H.npnct11.log H.npnct11.log -5.547032e-03 0
## H.npnct22.log H.npnct22.log -5.547032e-03 0
## S.npnct02.log S.npnct02.log -5.547032e-03 0
## S.npnct11.log S.npnct11.log -5.547032e-03 0
## A.npnct11.log A.npnct11.log -5.547032e-03 0
## A.npnct27.log A.npnct27.log -5.547032e-03 0
## A.npnct09.log A.npnct09.log -4.775988e-03 0
## S.npnct09.log S.npnct09.log -3.986882e-03 0
## A.npnct08.log A.npnct08.log -3.258100e-03 0
## H.T.time H.T.time -2.527450e-03 0
## S.npnct08.log S.npnct08.log -2.413868e-03 0
## S.T.presid S.T.presid -2.079562e-03 0
## A.T.presid A.T.presid -1.789086e-03 0
## S.npnct17.log S.npnct17.log -1.587454e-03 0
## A.npnct17.log A.npnct17.log -1.587454e-03 0
## H.T.take H.T.take -8.582583e-04 0
## H.npnct26.log H.npnct26.log -9.890046e-19 0
## S.npnct26.log S.npnct26.log -9.890046e-19 0
## A.npnct26.log A.npnct26.log -9.890046e-19 0
## H.has.http H.has.http NA 0
## H.npnct10.log H.npnct10.log NA 0
## H.npnct18.log H.npnct18.log NA 0
## H.npnct19.log H.npnct19.log NA 0
## H.npnct20.log H.npnct20.log NA 0
## H.npnct23.log H.npnct23.log NA 0
## H.npnct24.log H.npnct24.log NA 0
## H.npnct25.log H.npnct25.log NA 0
## H.npnct27.log H.npnct27.log NA 0
## H.npnct28.log H.npnct28.log NA 0
## H.npnct29.log H.npnct29.log NA 0
## H.npnct31.log H.npnct31.log NA 0
## H.npnct32.log H.npnct32.log NA 0
## S.has.http S.has.http NA 0
## S.npnct05.log S.npnct05.log NA 0
## S.npnct10.log S.npnct10.log NA 0
## S.npnct18.log S.npnct18.log NA 0
## S.npnct19.log S.npnct19.log NA 0
## S.npnct20.log S.npnct20.log NA 0
## S.npnct24.log S.npnct24.log NA 0
## S.npnct27.log S.npnct27.log NA 0
## S.npnct28.log S.npnct28.log NA 0
## S.npnct29.log S.npnct29.log NA 0
## S.npnct31.log S.npnct31.log NA 0
## S.npnct32.log S.npnct32.log NA 0
## A.npnct05.log A.npnct05.log NA 0
## A.npnct10.log A.npnct10.log NA 0
## A.npnct24.log A.npnct24.log NA 0
## A.npnct28.log A.npnct28.log NA 0
## A.npnct29.log A.npnct29.log NA 0
## A.npnct31.log A.npnct31.log NA 0
## A.npnct32.log A.npnct32.log NA 0
## clusterid clusterid NA 0
## PubDate.year.fctr PubDate.year.fctr NA 0
## cor.y.abs
## Popular 1.000000e+00
## A.nuppr.log 2.720962e-01
## S.nuppr.log 2.718459e-01
## WordCount.log 2.656836e-01
## WordCount 2.575265e-01
## S.nwrds.unq.log 2.507969e-01
## A.nwrds.unq.log 2.506012e-01
## S.nchrs.log 2.246930e-01
## A.nchrs.log 2.245488e-01
## H.nwrds.unq.log 2.044964e-01
## H.nchrs.log 1.710624e-01
## H.nwrds.log 1.410282e-01
## S.nwrds.log 1.359149e-01
## PubDate.hour.fctr 1.354368e-01
## A.nwrds.log 1.354108e-01
## H.npnct21.log 1.283641e-01
## H.nuppr.log 1.278085e-01
## A.ndgts.log 1.249484e-01
## S.ndgts.log 1.242046e-01
## H.ndgts.log 1.196633e-01
## PubDate.wkend 1.067288e-01
## A.npnct12.log 9.183870e-02
## S.npnct12.log 9.158156e-02
## H.npnct30.log 8.917338e-02
## S.T.week 8.503373e-02
## A.T.week 8.492895e-02
## S.T.fashion 8.419711e-02
## A.T.fashion 8.419345e-02
## H.npnct16.log 8.273237e-02
## H.T.fashion 7.947505e-02
## H.has.year.colon 7.842875e-02
## S.T.intern 6.953750e-02
## A.T.intern 6.949870e-02
## A.npnct16.log 6.893301e-02
## H.T.week 6.812724e-02
## S.npnct16.log 6.770952e-02
## H.T.X2015 6.601141e-02
## H.T.daili 6.299948e-02
## S.npnct04.log 6.294642e-02
## A.npnct04.log 6.294642e-02
## H.T.report 6.244050e-02
## H.npnct15.log 6.158577e-02
## H.T.day 6.033488e-02
## H.T.springsumm 5.943248e-02
## H.T.today 5.831308e-02
## S.T.newyork 5.712853e-02
## A.T.newyork 5.706083e-02
## H.T.newyork 5.650839e-02
## S.npnct21.log 5.503894e-02
## A.npnct21.log 5.482747e-02
## S.T.articl 5.446201e-02
## A.T.articl 5.445243e-02
## PubDate.last10 5.398093e-02
## H.npnct08.log 5.375262e-02
## H.npnct09.log 5.375262e-02
## S.npnct14.log 5.332519e-02
## A.T.time 5.313395e-02
## S.T.time 5.303654e-02
## H.npnct04.log 5.126277e-02
## S.T.share 5.070234e-02
## A.T.share 5.070234e-02
## A.npnct14.log 4.999563e-02
## PubDate.last10.log 4.931702e-02
## H.T.busi 4.901905e-02
## H.T.morn 4.838380e-02
## S.T.compani 4.764341e-02
## A.T.compani 4.751471e-02
## S.T.report 4.746920e-02
## A.T.report 4.741555e-02
## S.T.year 4.729011e-02
## A.T.year 4.721236e-02
## PubDate.last1.log 4.635751e-02
## S.T.first 4.617532e-02
## A.T.first 4.603341e-02
## H.T.X2014 4.497745e-02
## H.T.first 4.458885e-02
## H.T.news 4.415284e-02
## A.npnct30.log 4.373349e-02
## S.npnct30.log 4.370037e-02
## H.T.new 4.327803e-02
## A.T.day 4.196599e-02
## A.T.show 4.196129e-02
## S.T.show 4.193803e-02
## S.T.day 4.188671e-02
## H.T.pictur 3.993172e-02
## PubDate.last100 3.989229e-02
## PubDate.wkday.fctr 3.980129e-02
## A.T.make 3.965722e-02
## S.T.make 3.959645e-02
## S.T.will 3.892267e-02
## A.T.will 3.887937e-02
## A.npnct13.log 3.760012e-02
## S.npnct13.log 3.638891e-02
## PubDate.last1 3.592267e-02
## PubDate.minute.fctr 3.407385e-02
## H.T.art 3.280483e-02
## H.npnct06.log 3.190718e-02
## H.T.china 3.144808e-02
## A.T.can 3.127063e-02
## S.npnct01.log 3.093101e-02
## A.npnct01.log 3.093101e-02
## H.T.polit 3.062866e-02
## S.T.can 3.049697e-02
## H.npnct17.log 3.039622e-02
## H.T.billion 2.949817e-02
## A.T.new 2.782876e-02
## S.T.new 2.769558e-02
## S.npnct23.log 2.760321e-02
## S.npnct25.log 2.760321e-02
## H.has.ebola 2.588140e-02
## H.T.deal 2.559418e-02
## H.npnct14.log 2.524770e-02
## A.npnct15.log 2.407715e-02
## S.npnct06.log 2.389145e-02
## A.npnct06.log 2.389145e-02
## A.T.take 2.282555e-02
## S.T.take 2.275732e-02
## H.npnct01.log 2.271577e-02
## S.npnct15.log 2.121844e-02
## H.T.test 2.065255e-02
## H.npnct02.log 2.001851e-02
## S.npnct22.log 1.923169e-02
## A.npnct22.log 1.923169e-02
## PubDate.month.fctr 1.914874e-02
## A.T.said 1.839871e-02
## S.T.said 1.826884e-02
## S.has.year.colon 1.755336e-02
## A.has.year.colon 1.755336e-02
## PubDate.POSIX 1.568326e-02
## PubDate.zoo 1.568326e-02
## A.npnct23.log 1.537569e-02
## A.npnct25.log 1.537569e-02
## A.npnct02.log 1.451467e-02
## A.npnct18.log 1.451467e-02
## A.npnct20.log 1.451467e-02
## H.T.make 1.430572e-02
## H.T.big 1.390748e-02
## A.has.http 1.359260e-02
## A.npnct03.log 1.359260e-02
## H.npnct12.log 1.333613e-02
## H.npnct13.log 1.305305e-02
## A.npnct19.log 1.271661e-02
## S.npnct03.log 1.240734e-02
## myCategory.fctr 1.234541e-02
## S.npnct07.log 1.214357e-02
## A.npnct07.log 1.214357e-02
## H.npnct07.log 1.201741e-02
## PubDate.second.fctr 1.187946e-02
## UniqueID 1.182492e-02
## PubDate.date.fctr 1.164756e-02
## A.T.one 1.081694e-02
## S.T.one 1.080534e-02
## A.T.state 1.020706e-02
## S.T.state 1.012205e-02
## H.T.bank 9.989139e-03
## H.T.say 9.960773e-03
## H.T.obama 9.907543e-03
## H.npnct05.log 9.653967e-03
## H.npnct03.log 9.533020e-03
## .rnorm 8.244230e-03
## PubDate.last100.log 7.663322e-03
## H.npnct11.log 5.547032e-03
## H.npnct22.log 5.547032e-03
## S.npnct02.log 5.547032e-03
## S.npnct11.log 5.547032e-03
## A.npnct11.log 5.547032e-03
## A.npnct27.log 5.547032e-03
## A.npnct09.log 4.775988e-03
## S.npnct09.log 3.986882e-03
## A.npnct08.log 3.258100e-03
## H.T.time 2.527450e-03
## S.npnct08.log 2.413868e-03
## S.T.presid 2.079562e-03
## A.T.presid 1.789086e-03
## S.npnct17.log 1.587454e-03
## A.npnct17.log 1.587454e-03
## H.T.take 8.582583e-04
## H.npnct26.log 9.890046e-19
## S.npnct26.log 9.890046e-19
## A.npnct26.log 9.890046e-19
## H.has.http NA
## H.npnct10.log NA
## H.npnct18.log NA
## H.npnct19.log NA
## H.npnct20.log NA
## H.npnct23.log NA
## H.npnct24.log NA
## H.npnct25.log NA
## H.npnct27.log NA
## H.npnct28.log NA
## H.npnct29.log NA
## H.npnct31.log NA
## H.npnct32.log NA
## S.has.http NA
## S.npnct05.log NA
## S.npnct10.log NA
## S.npnct18.log NA
## S.npnct19.log NA
## S.npnct20.log NA
## S.npnct24.log NA
## S.npnct27.log NA
## S.npnct28.log NA
## S.npnct29.log NA
## S.npnct31.log NA
## S.npnct32.log NA
## A.npnct05.log NA
## A.npnct10.log NA
## A.npnct24.log NA
## A.npnct28.log NA
## A.npnct29.log NA
## A.npnct31.log NA
## A.npnct32.log NA
## clusterid NA
## PubDate.year.fctr NA# sav_feats_df <- glb_feats_df
print(glb_feats_df <- orderBy(~-cor.y,
myfind_cor_features(feats_df=glb_feats_df, entity_df=glb_trnent_df,
rsp_var=glb_rsp_var)))## Loading required package: caret
##
## Attaching package: 'caret'
##
## The following object is masked from 'package:survival':
##
## cluster## [1] "cor(A.has.year.colon, S.has.year.colon)=1.0000"
## [1] "cor(Popular.fctr, A.has.year.colon)=-0.0176"
## [1] "cor(Popular.fctr, S.has.year.colon)=-0.0176"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.has.year.colon as highly correlated with
## A.has.year.colon## [1] "cor(A.npnct01.log, S.npnct01.log)=1.0000"
## [1] "cor(Popular.fctr, A.npnct01.log)=0.0309"
## [1] "cor(Popular.fctr, S.npnct01.log)=0.0309"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct01.log as highly correlated with
## A.npnct01.log## [1] "cor(A.npnct04.log, S.npnct04.log)=1.0000"
## [1] "cor(Popular.fctr, A.npnct04.log)=-0.0629"
## [1] "cor(Popular.fctr, S.npnct04.log)=-0.0629"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct04.log as highly correlated with
## A.npnct04.log## [1] "cor(A.npnct06.log, S.npnct06.log)=1.0000"
## [1] "cor(Popular.fctr, A.npnct06.log)=-0.0239"
## [1] "cor(Popular.fctr, S.npnct06.log)=-0.0239"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct06.log as highly correlated with
## A.npnct06.log## [1] "cor(A.npnct07.log, S.npnct07.log)=1.0000"
## [1] "cor(Popular.fctr, A.npnct07.log)=-0.0121"
## [1] "cor(Popular.fctr, S.npnct07.log)=-0.0121"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct07.log as highly correlated with
## A.npnct07.log## [1] "cor(A.npnct18.log, A.npnct20.log)=1.0000"
## [1] "cor(Popular.fctr, A.npnct18.log)=-0.0145"
## [1] "cor(Popular.fctr, A.npnct20.log)=-0.0145"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.npnct20.log as highly correlated with
## A.npnct18.log## [1] "cor(A.npnct22.log, S.npnct22.log)=1.0000"
## [1] "cor(Popular.fctr, A.npnct22.log)=-0.0192"
## [1] "cor(Popular.fctr, S.npnct22.log)=-0.0192"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct22.log as highly correlated with
## A.npnct22.log## [1] "cor(A.npnct23.log, A.npnct25.log)=1.0000"
## [1] "cor(Popular.fctr, A.npnct23.log)=0.0154"
## [1] "cor(Popular.fctr, A.npnct25.log)=0.0154"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.npnct25.log as highly correlated with
## A.npnct23.log## [1] "cor(A.T.share, S.T.share)=1.0000"
## [1] "cor(Popular.fctr, A.T.share)=-0.0507"
## [1] "cor(Popular.fctr, S.T.share)=-0.0507"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.share as highly correlated with A.T.share## [1] "cor(H.npnct08.log, H.npnct09.log)=1.0000"
## [1] "cor(Popular.fctr, H.npnct08.log)=0.0538"
## [1] "cor(Popular.fctr, H.npnct09.log)=0.0538"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.npnct09.log as highly correlated with
## H.npnct08.log## [1] "cor(S.npnct23.log, S.npnct25.log)=1.0000"
## [1] "cor(Popular.fctr, S.npnct23.log)=0.0276"
## [1] "cor(Popular.fctr, S.npnct25.log)=0.0276"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct25.log as highly correlated with
## S.npnct23.log## [1] "cor(A.T.fashion, S.T.fashion)=1.0000"
## [1] "cor(Popular.fctr, A.T.fashion)=-0.0842"
## [1] "cor(Popular.fctr, S.T.fashion)=-0.0842"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.fashion as highly correlated with
## S.T.fashion## [1] "cor(A.T.articl, S.T.articl)=1.0000"
## [1] "cor(Popular.fctr, A.T.articl)=-0.0545"
## [1] "cor(Popular.fctr, S.T.articl)=-0.0545"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.articl as highly correlated with
## S.T.articl## [1] "cor(A.T.show, S.T.show)=1.0000"
## [1] "cor(Popular.fctr, A.T.show)=-0.0420"
## [1] "cor(Popular.fctr, S.T.show)=-0.0419"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.show as highly correlated with A.T.show## [1] "cor(A.T.make, S.T.make)=1.0000"
## [1] "cor(Popular.fctr, A.T.make)=0.0397"
## [1] "cor(Popular.fctr, S.T.make)=0.0396"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.make as highly correlated with A.T.make## [1] "cor(A.T.newyork, S.T.newyork)=1.0000"
## [1] "cor(Popular.fctr, A.T.newyork)=-0.0571"
## [1] "cor(Popular.fctr, S.T.newyork)=-0.0571"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.newyork as highly correlated with
## S.T.newyork## [1] "cor(A.T.intern, S.T.intern)=1.0000"
## [1] "cor(Popular.fctr, A.T.intern)=-0.0695"
## [1] "cor(Popular.fctr, S.T.intern)=-0.0695"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.intern as highly correlated with
## S.T.intern## [1] "cor(A.T.report, S.T.report)=1.0000"
## [1] "cor(Popular.fctr, A.T.report)=-0.0474"
## [1] "cor(Popular.fctr, S.T.report)=-0.0475"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.report as highly correlated with
## S.T.report## [1] "cor(A.T.year, S.T.year)=1.0000"
## [1] "cor(Popular.fctr, A.T.year)=-0.0472"
## [1] "cor(Popular.fctr, S.T.year)=-0.0473"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.year as highly correlated with S.T.year## [1] "cor(A.T.week, S.T.week)=0.9999"
## [1] "cor(Popular.fctr, A.T.week)=-0.0849"
## [1] "cor(Popular.fctr, S.T.week)=-0.0850"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.week as highly correlated with S.T.week## [1] "cor(A.T.said, S.T.said)=0.9999"
## [1] "cor(Popular.fctr, A.T.said)=0.0184"
## [1] "cor(Popular.fctr, S.T.said)=0.0183"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.said as highly correlated with A.T.said## [1] "cor(A.T.compani, S.T.compani)=0.9999"
## [1] "cor(Popular.fctr, A.T.compani)=-0.0475"
## [1] "cor(Popular.fctr, S.T.compani)=-0.0476"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.compani as highly correlated with
## S.T.compani## [1] "cor(A.T.first, S.T.first)=0.9998"
## [1] "cor(Popular.fctr, A.T.first)=-0.0460"
## [1] "cor(Popular.fctr, S.T.first)=-0.0462"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.first as highly correlated with S.T.first## [1] "cor(A.T.time, S.T.time)=0.9998"
## [1] "cor(Popular.fctr, A.T.time)=-0.0531"
## [1] "cor(Popular.fctr, S.T.time)=-0.0530"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.time as highly correlated with A.T.time## [1] "cor(A.npnct12.log, S.npnct12.log)=0.9997"
## [1] "cor(Popular.fctr, A.npnct12.log)=-0.0918"
## [1] "cor(Popular.fctr, S.npnct12.log)=-0.0916"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct12.log as highly correlated with
## A.npnct12.log## [1] "cor(A.nwrds.log, S.nwrds.log)=0.9997"
## [1] "cor(Popular.fctr, A.nwrds.log)=0.1354"
## [1] "cor(Popular.fctr, S.nwrds.log)=0.1359"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.nwrds.log as highly correlated with
## S.nwrds.log## [1] "cor(A.T.one, S.T.one)=0.9997"
## [1] "cor(Popular.fctr, A.T.one)=0.0108"
## [1] "cor(Popular.fctr, S.T.one)=0.0108"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.one as highly correlated with A.T.one## [1] "cor(A.T.state, S.T.state)=0.9997"
## [1] "cor(Popular.fctr, A.T.state)=0.0102"
## [1] "cor(Popular.fctr, S.T.state)=0.0101"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.state as highly correlated with A.T.state## [1] "cor(A.T.take, S.T.take)=0.9997"
## [1] "cor(Popular.fctr, A.T.take)=-0.0228"
## [1] "cor(Popular.fctr, S.T.take)=-0.0228"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.take as highly correlated with A.T.take## [1] "cor(A.T.new, S.T.new)=0.9997"
## [1] "cor(Popular.fctr, A.T.new)=-0.0278"
## [1] "cor(Popular.fctr, S.T.new)=-0.0277"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.new as highly correlated with A.T.new## [1] "cor(A.T.can, S.T.can)=0.9996"
## [1] "cor(Popular.fctr, A.T.can)=0.0313"
## [1] "cor(Popular.fctr, S.T.can)=0.0305"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.can as highly correlated with A.T.can## [1] "cor(A.T.will, S.T.will)=0.9996"
## [1] "cor(Popular.fctr, A.T.will)=-0.0389"
## [1] "cor(Popular.fctr, S.T.will)=-0.0389"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.T.will as highly correlated with S.T.will## [1] "cor(A.T.day, S.T.day)=0.9996"
## [1] "cor(Popular.fctr, A.T.day)=-0.0420"
## [1] "cor(Popular.fctr, S.T.day)=-0.0419"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.day as highly correlated with A.T.day## [1] "cor(A.nuppr.log, S.nuppr.log)=0.9991"
## [1] "cor(Popular.fctr, A.nuppr.log)=-0.2721"
## [1] "cor(Popular.fctr, S.nuppr.log)=-0.2718"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.nuppr.log as highly correlated with
## A.nuppr.log## [1] "cor(A.npnct30.log, S.npnct30.log)=0.9989"
## [1] "cor(Popular.fctr, A.npnct30.log)=-0.0437"
## [1] "cor(Popular.fctr, S.npnct30.log)=-0.0437"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct30.log as highly correlated with
## A.npnct30.log## [1] "cor(A.nwrds.unq.log, S.nwrds.unq.log)=0.9989"
## [1] "cor(Popular.fctr, A.nwrds.unq.log)=-0.2506"
## [1] "cor(Popular.fctr, S.nwrds.unq.log)=-0.2508"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.nwrds.unq.log as highly correlated with
## S.nwrds.unq.log## [1] "cor(A.nchrs.log, S.nchrs.log)=0.9986"
## [1] "cor(Popular.fctr, A.nchrs.log)=-0.2245"
## [1] "cor(Popular.fctr, S.nchrs.log)=-0.2247"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.nchrs.log as highly correlated with
## S.nchrs.log## [1] "cor(A.npnct21.log, S.npnct21.log)=0.9957"
## [1] "cor(Popular.fctr, A.npnct21.log)=0.0548"
## [1] "cor(Popular.fctr, S.npnct21.log)=0.0550"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.npnct21.log as highly correlated with
## S.npnct21.log## [1] "cor(A.ndgts.log, S.ndgts.log)=0.9955"
## [1] "cor(Popular.fctr, A.ndgts.log)=-0.1249"
## [1] "cor(Popular.fctr, S.ndgts.log)=-0.1242"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.ndgts.log as highly correlated with
## A.ndgts.log## [1] "cor(A.npnct13.log, S.npnct13.log)=0.9935"
## [1] "cor(Popular.fctr, A.npnct13.log)=-0.0376"
## [1] "cor(Popular.fctr, S.npnct13.log)=-0.0364"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct13.log as highly correlated with
## A.npnct13.log## [1] "cor(A.npnct16.log, S.npnct16.log)=0.9917"
## [1] "cor(Popular.fctr, A.npnct16.log)=-0.0689"
## [1] "cor(Popular.fctr, S.npnct16.log)=-0.0677"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct16.log as highly correlated with
## A.npnct16.log## [1] "cor(A.npnct14.log, S.npnct14.log)=0.9795"
## [1] "cor(Popular.fctr, A.npnct14.log)=-0.0500"
## [1] "cor(Popular.fctr, S.npnct14.log)=-0.0533"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.npnct14.log as highly correlated with
## S.npnct14.log## [1] "cor(H.npnct15.log, H.T.springsumm)=0.9650"
## [1] "cor(Popular.fctr, H.npnct15.log)=-0.0616"
## [1] "cor(Popular.fctr, H.T.springsumm)=-0.0594"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.T.springsumm as highly correlated with
## H.npnct15.log## [1] "cor(S.nchrs.log, S.nwrds.unq.log)=0.9543"
## [1] "cor(Popular.fctr, S.nchrs.log)=-0.2247"
## [1] "cor(Popular.fctr, S.nwrds.unq.log)=-0.2508"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.nchrs.log as highly correlated with
## S.nwrds.unq.log## [1] "cor(A.has.http, A.npnct19.log)=0.9356"
## [1] "cor(Popular.fctr, A.has.http)=-0.0136"
## [1] "cor(Popular.fctr, A.npnct19.log)=-0.0127"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.npnct19.log as highly correlated with
## A.has.http## [1] "cor(A.has.http, A.npnct02.log)=0.9247"
## [1] "cor(Popular.fctr, A.has.http)=-0.0136"
## [1] "cor(Popular.fctr, A.npnct02.log)=-0.0145"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.has.http as highly correlated with
## A.npnct02.log## [1] "cor(A.npnct03.log, S.npnct03.log)=0.9128"
## [1] "cor(Popular.fctr, A.npnct03.log)=-0.0136"
## [1] "cor(Popular.fctr, S.npnct03.log)=-0.0124"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.npnct03.log as highly correlated with
## A.npnct03.log## [1] "cor(H.nchrs.log, H.nwrds.unq.log)=0.8881"
## [1] "cor(Popular.fctr, H.nchrs.log)=-0.1711"
## [1] "cor(Popular.fctr, H.nwrds.unq.log)=-0.2045"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.nchrs.log as highly correlated with
## H.nwrds.unq.log## [1] "cor(H.npnct15.log, H.T.X2015)=0.8848"
## [1] "cor(Popular.fctr, H.npnct15.log)=-0.0616"
## [1] "cor(Popular.fctr, H.T.X2015)=-0.0660"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.npnct15.log as highly correlated with
## H.T.X2015## [1] "cor(A.npnct02.log, A.npnct18.log)=0.8771"
## [1] "cor(Popular.fctr, A.npnct02.log)=-0.0145"
## [1] "cor(Popular.fctr, A.npnct18.log)=-0.0145"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.npnct18.log as highly correlated with
## A.npnct02.log## [1] "cor(A.npnct30.log, H.T.morn)=0.8327"
## [1] "cor(Popular.fctr, A.npnct30.log)=-0.0437"
## [1] "cor(Popular.fctr, H.T.morn)=-0.0484"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.npnct30.log as highly correlated with
## H.T.morn## [1] "cor(H.nuppr.log, H.nwrds.unq.log)=0.8288"
## [1] "cor(Popular.fctr, H.nuppr.log)=-0.1278"
## [1] "cor(Popular.fctr, H.nwrds.unq.log)=-0.2045"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.nuppr.log as highly correlated with
## H.nwrds.unq.log## [1] "cor(H.T.daili, H.T.report)=0.8265"
## [1] "cor(Popular.fctr, H.T.daili)=-0.0630"
## [1] "cor(Popular.fctr, H.T.report)=-0.0624"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.T.report as highly correlated with H.T.daili## [1] "cor(H.npnct06.log, H.npnct17.log)=0.8106"
## [1] "cor(Popular.fctr, H.npnct06.log)=0.0319"
## [1] "cor(Popular.fctr, H.npnct17.log)=0.0304"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.npnct17.log as highly correlated with
## H.npnct06.log## [1] "cor(H.has.year.colon, S.T.intern)=0.8073"
## [1] "cor(Popular.fctr, H.has.year.colon)=-0.0784"
## [1] "cor(Popular.fctr, S.T.intern)=-0.0695"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified S.T.intern as highly correlated with
## H.has.year.colon## [1] "cor(A.npnct23.log, S.npnct23.log)=0.7461"
## [1] "cor(Popular.fctr, A.npnct23.log)=0.0154"
## [1] "cor(Popular.fctr, S.npnct23.log)=0.0276"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.npnct23.log as highly correlated with
## S.npnct23.log## [1] "cor(H.T.polit, H.T.today)=0.7371"
## [1] "cor(Popular.fctr, H.T.polit)=-0.0306"
## [1] "cor(Popular.fctr, H.T.today)=-0.0583"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.T.polit as highly correlated with H.T.today## [1] "cor(A.npnct02.log, A.npnct15.log)=0.7324"
## [1] "cor(Popular.fctr, A.npnct02.log)=-0.0145"
## [1] "cor(Popular.fctr, A.npnct15.log)=-0.0241"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.npnct02.log as highly correlated with
## A.npnct15.log## [1] "cor(H.T.X2015, S.T.fashion)=0.7292"
## [1] "cor(Popular.fctr, H.T.X2015)=-0.0660"
## [1] "cor(Popular.fctr, S.T.fashion)=-0.0842"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.T.X2015 as highly correlated with
## S.T.fashion## [1] "cor(H.npnct04.log, H.T.billion)=0.7010"
## [1] "cor(Popular.fctr, H.npnct04.log)=-0.0513"
## [1] "cor(Popular.fctr, H.T.billion)=-0.0295"## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified H.T.billion as highly correlated with
## H.npnct04.log## id cor.y exclude.as.feat
## Popular Popular 1.000000e+00 1
## WordCount.log WordCount.log 2.656836e-01 0
## WordCount WordCount 2.575265e-01 1
## H.nwrds.log H.nwrds.log 1.410282e-01 0
## S.nwrds.log S.nwrds.log 1.359149e-01 0
## PubDate.hour.fctr PubDate.hour.fctr 1.354368e-01 0
## A.nwrds.log A.nwrds.log 1.354108e-01 0
## H.npnct21.log H.npnct21.log 1.283641e-01 0
## PubDate.wkend PubDate.wkend 1.067288e-01 0
## S.npnct21.log S.npnct21.log 5.503894e-02 0
## A.npnct21.log A.npnct21.log 5.482747e-02 0
## PubDate.last10 PubDate.last10 5.398093e-02 1
## H.npnct08.log H.npnct08.log 5.375262e-02 0
## H.npnct09.log H.npnct09.log 5.375262e-02 0
## PubDate.last10.log PubDate.last10.log 4.931702e-02 0
## PubDate.last1.log PubDate.last1.log 4.635751e-02 0
## PubDate.last100 PubDate.last100 3.989229e-02 1
## A.T.make A.T.make 3.965722e-02 0
## S.T.make S.T.make 3.959645e-02 0
## PubDate.last1 PubDate.last1 3.592267e-02 1
## H.npnct06.log H.npnct06.log 3.190718e-02 0
## A.T.can A.T.can 3.127063e-02 0
## A.npnct01.log A.npnct01.log 3.093101e-02 0
## S.npnct01.log S.npnct01.log 3.093101e-02 0
## S.T.can S.T.can 3.049697e-02 0
## H.npnct17.log H.npnct17.log 3.039622e-02 0
## S.npnct23.log S.npnct23.log 2.760321e-02 0
## S.npnct25.log S.npnct25.log 2.760321e-02 0
## H.has.ebola H.has.ebola 2.588140e-02 0
## H.npnct01.log H.npnct01.log 2.271577e-02 0
## PubDate.month.fctr PubDate.month.fctr 1.914874e-02 1
## A.T.said A.T.said 1.839871e-02 0
## S.T.said S.T.said 1.826884e-02 0
## PubDate.POSIX PubDate.POSIX 1.568326e-02 1
## PubDate.zoo PubDate.zoo 1.568326e-02 1
## A.npnct23.log A.npnct23.log 1.537569e-02 0
## A.npnct25.log A.npnct25.log 1.537569e-02 0
## H.T.make H.T.make 1.430572e-02 0
## H.npnct12.log H.npnct12.log 1.333613e-02 0
## myCategory.fctr myCategory.fctr 1.234541e-02 0
## UniqueID UniqueID 1.182492e-02 1
## A.T.one A.T.one 1.081694e-02 0
## S.T.one S.T.one 1.080534e-02 0
## A.T.state A.T.state 1.020706e-02 0
## S.T.state S.T.state 1.012205e-02 0
## H.npnct03.log H.npnct03.log 9.533020e-03 0
## A.npnct26.log A.npnct26.log -9.890046e-19 0
## H.npnct26.log H.npnct26.log -9.890046e-19 0
## S.npnct26.log S.npnct26.log -9.890046e-19 0
## H.T.take H.T.take -8.582583e-04 0
## A.npnct17.log A.npnct17.log -1.587454e-03 0
## S.npnct17.log S.npnct17.log -1.587454e-03 0
## A.T.presid A.T.presid -1.789086e-03 0
## S.T.presid S.T.presid -2.079562e-03 0
## S.npnct08.log S.npnct08.log -2.413868e-03 0
## H.T.time H.T.time -2.527450e-03 0
## A.npnct08.log A.npnct08.log -3.258100e-03 0
## S.npnct09.log S.npnct09.log -3.986882e-03 0
## A.npnct09.log A.npnct09.log -4.775988e-03 0
## A.npnct27.log A.npnct27.log -5.547032e-03 0
## A.npnct11.log A.npnct11.log -5.547032e-03 0
## H.npnct11.log H.npnct11.log -5.547032e-03 0
## H.npnct22.log H.npnct22.log -5.547032e-03 0
## S.npnct02.log S.npnct02.log -5.547032e-03 0
## S.npnct11.log S.npnct11.log -5.547032e-03 0
## PubDate.last100.log PubDate.last100.log -7.663322e-03 0
## .rnorm .rnorm -8.244230e-03 0
## H.npnct05.log H.npnct05.log -9.653967e-03 0
## H.T.obama H.T.obama -9.907543e-03 0
## H.T.say H.T.say -9.960773e-03 0
## H.T.bank H.T.bank -9.989139e-03 0
## PubDate.date.fctr PubDate.date.fctr -1.164756e-02 0
## PubDate.second.fctr PubDate.second.fctr -1.187946e-02 0
## H.npnct07.log H.npnct07.log -1.201741e-02 0
## A.npnct07.log A.npnct07.log -1.214357e-02 0
## S.npnct07.log S.npnct07.log -1.214357e-02 0
## S.npnct03.log S.npnct03.log -1.240734e-02 0
## A.npnct19.log A.npnct19.log -1.271661e-02 0
## H.npnct13.log H.npnct13.log -1.305305e-02 0
## A.has.http A.has.http -1.359260e-02 0
## A.npnct03.log A.npnct03.log -1.359260e-02 0
## H.T.big H.T.big -1.390748e-02 0
## A.npnct02.log A.npnct02.log -1.451467e-02 0
## A.npnct18.log A.npnct18.log -1.451467e-02 0
## A.npnct20.log A.npnct20.log -1.451467e-02 0
## A.has.year.colon A.has.year.colon -1.755336e-02 0
## S.has.year.colon S.has.year.colon -1.755336e-02 0
## A.npnct22.log A.npnct22.log -1.923169e-02 0
## S.npnct22.log S.npnct22.log -1.923169e-02 0
## H.npnct02.log H.npnct02.log -2.001851e-02 0
## H.T.test H.T.test -2.065255e-02 0
## S.npnct15.log S.npnct15.log -2.121844e-02 0
## S.T.take S.T.take -2.275732e-02 0
## A.T.take A.T.take -2.282555e-02 0
## A.npnct06.log A.npnct06.log -2.389145e-02 0
## S.npnct06.log S.npnct06.log -2.389145e-02 0
## A.npnct15.log A.npnct15.log -2.407715e-02 0
## H.npnct14.log H.npnct14.log -2.524770e-02 0
## H.T.deal H.T.deal -2.559418e-02 0
## S.T.new S.T.new -2.769558e-02 0
## A.T.new A.T.new -2.782876e-02 0
## H.T.billion H.T.billion -2.949817e-02 0
## H.T.polit H.T.polit -3.062866e-02 0
## H.T.china H.T.china -3.144808e-02 0
## H.T.art H.T.art -3.280483e-02 0
## PubDate.minute.fctr PubDate.minute.fctr -3.407385e-02 0
## S.npnct13.log S.npnct13.log -3.638891e-02 0
## A.npnct13.log A.npnct13.log -3.760012e-02 0
## A.T.will A.T.will -3.887937e-02 0
## S.T.will S.T.will -3.892267e-02 0
## PubDate.wkday.fctr PubDate.wkday.fctr -3.980129e-02 0
## H.T.pictur H.T.pictur -3.993172e-02 0
## S.T.day S.T.day -4.188671e-02 0
## S.T.show S.T.show -4.193803e-02 0
## A.T.show A.T.show -4.196129e-02 0
## A.T.day A.T.day -4.196599e-02 0
## H.T.new H.T.new -4.327803e-02 0
## S.npnct30.log S.npnct30.log -4.370037e-02 0
## A.npnct30.log A.npnct30.log -4.373349e-02 0
## H.T.news H.T.news -4.415284e-02 0
## H.T.first H.T.first -4.458885e-02 0
## H.T.X2014 H.T.X2014 -4.497745e-02 0
## A.T.first A.T.first -4.603341e-02 0
## S.T.first S.T.first -4.617532e-02 0
## A.T.year A.T.year -4.721236e-02 0
## S.T.year S.T.year -4.729011e-02 0
## A.T.report A.T.report -4.741555e-02 0
## S.T.report S.T.report -4.746920e-02 0
## A.T.compani A.T.compani -4.751471e-02 0
## S.T.compani S.T.compani -4.764341e-02 0
## H.T.morn H.T.morn -4.838380e-02 0
## H.T.busi H.T.busi -4.901905e-02 0
## A.npnct14.log A.npnct14.log -4.999563e-02 0
## A.T.share A.T.share -5.070234e-02 0
## S.T.share S.T.share -5.070234e-02 0
## H.npnct04.log H.npnct04.log -5.126277e-02 0
## S.T.time S.T.time -5.303654e-02 0
## A.T.time A.T.time -5.313395e-02 0
## S.npnct14.log S.npnct14.log -5.332519e-02 0
## A.T.articl A.T.articl -5.445243e-02 0
## S.T.articl S.T.articl -5.446201e-02 0
## H.T.newyork H.T.newyork -5.650839e-02 0
## A.T.newyork A.T.newyork -5.706083e-02 0
## S.T.newyork S.T.newyork -5.712853e-02 0
## H.T.today H.T.today -5.831308e-02 0
## H.T.springsumm H.T.springsumm -5.943248e-02 0
## H.T.day H.T.day -6.033488e-02 0
## H.npnct15.log H.npnct15.log -6.158577e-02 0
## H.T.report H.T.report -6.244050e-02 0
## A.npnct04.log A.npnct04.log -6.294642e-02 0
## S.npnct04.log S.npnct04.log -6.294642e-02 0
## H.T.daili H.T.daili -6.299948e-02 0
## H.T.X2015 H.T.X2015 -6.601141e-02 0
## S.npnct16.log S.npnct16.log -6.770952e-02 0
## H.T.week H.T.week -6.812724e-02 0
## A.npnct16.log A.npnct16.log -6.893301e-02 0
## A.T.intern A.T.intern -6.949870e-02 0
## S.T.intern S.T.intern -6.953750e-02 0
## H.has.year.colon H.has.year.colon -7.842875e-02 0
## H.T.fashion H.T.fashion -7.947505e-02 0
## H.npnct16.log H.npnct16.log -8.273237e-02 0
## A.T.fashion A.T.fashion -8.419345e-02 0
## S.T.fashion S.T.fashion -8.419711e-02 0
## A.T.week A.T.week -8.492895e-02 0
## S.T.week S.T.week -8.503373e-02 0
## H.npnct30.log H.npnct30.log -8.917338e-02 0
## S.npnct12.log S.npnct12.log -9.158156e-02 0
## A.npnct12.log A.npnct12.log -9.183870e-02 0
## H.ndgts.log H.ndgts.log -1.196633e-01 0
## S.ndgts.log S.ndgts.log -1.242046e-01 0
## A.ndgts.log A.ndgts.log -1.249484e-01 0
## H.nuppr.log H.nuppr.log -1.278085e-01 0
## H.nchrs.log H.nchrs.log -1.710624e-01 0
## H.nwrds.unq.log H.nwrds.unq.log -2.044964e-01 0
## A.nchrs.log A.nchrs.log -2.245488e-01 0
## S.nchrs.log S.nchrs.log -2.246930e-01 0
## A.nwrds.unq.log A.nwrds.unq.log -2.506012e-01 0
## S.nwrds.unq.log S.nwrds.unq.log -2.507969e-01 0
## S.nuppr.log S.nuppr.log -2.718459e-01 0
## A.nuppr.log A.nuppr.log -2.720962e-01 0
## A.npnct05.log A.npnct05.log NA 0
## A.npnct10.log A.npnct10.log NA 0
## A.npnct24.log A.npnct24.log NA 0
## A.npnct28.log A.npnct28.log NA 0
## A.npnct29.log A.npnct29.log NA 0
## A.npnct31.log A.npnct31.log NA 0
## A.npnct32.log A.npnct32.log NA 0
## clusterid clusterid NA 0
## H.has.http H.has.http NA 0
## H.npnct10.log H.npnct10.log NA 0
## H.npnct18.log H.npnct18.log NA 0
## H.npnct19.log H.npnct19.log NA 0
## H.npnct20.log H.npnct20.log NA 0
## H.npnct23.log H.npnct23.log NA 0
## H.npnct24.log H.npnct24.log NA 0
## H.npnct25.log H.npnct25.log NA 0
## H.npnct27.log H.npnct27.log NA 0
## H.npnct28.log H.npnct28.log NA 0
## H.npnct29.log H.npnct29.log NA 0
## H.npnct31.log H.npnct31.log NA 0
## H.npnct32.log H.npnct32.log NA 0
## PubDate.year.fctr PubDate.year.fctr NA 0
## S.has.http S.has.http NA 0
## S.npnct05.log S.npnct05.log NA 0
## S.npnct10.log S.npnct10.log NA 0
## S.npnct18.log S.npnct18.log NA 0
## S.npnct19.log S.npnct19.log NA 0
## S.npnct20.log S.npnct20.log NA 0
## S.npnct24.log S.npnct24.log NA 0
## S.npnct27.log S.npnct27.log NA 0
## S.npnct28.log S.npnct28.log NA 0
## S.npnct29.log S.npnct29.log NA 0
## S.npnct31.log S.npnct31.log NA 0
## S.npnct32.log S.npnct32.log NA 0
## cor.y.abs cor.high.X freqRatio
## Popular 1.000000e+00 <NA> 4.976212
## WordCount.log 2.656836e-01 <NA> 1.300000
## WordCount 2.575265e-01 <NA> 2.315789
## H.nwrds.log 1.410282e-01 <NA> 1.127273
## S.nwrds.log 1.359149e-01 A.nwrds.log 2.583333
## PubDate.hour.fctr 1.354368e-01 <NA> 1.835040
## A.nwrds.log 1.354108e-01 <NA> 2.583333
## H.npnct21.log 1.283641e-01 <NA> 14.995098
## PubDate.wkend 1.067288e-01 <NA> 9.095827
## S.npnct21.log 5.503894e-02 A.npnct21.log 12.862366
## A.npnct21.log 5.482747e-02 <NA> 12.798715
## PubDate.last10 5.398093e-02 <NA> 1.666667
## H.npnct08.log 5.375262e-02 H.npnct09.log 111.620690
## H.npnct09.log 5.375262e-02 <NA> 111.620690
## PubDate.last10.log 4.931702e-02 <NA> 1.666667
## PubDate.last1.log 4.635751e-02 <NA> 1.142857
## PubDate.last100 3.989229e-02 <NA> 25.000000
## A.T.make 3.965722e-02 S.T.make 273.782609
## S.T.make 3.959645e-02 <NA> 273.782609
## PubDate.last1 3.592267e-02 <NA> 1.142857
## H.npnct06.log 3.190718e-02 H.npnct17.log 68.935484
## A.T.can 3.127063e-02 S.T.can 261.666667
## A.npnct01.log 3.093101e-02 S.npnct01.log 309.952381
## S.npnct01.log 3.093101e-02 <NA> 309.952381
## S.T.can 3.049697e-02 <NA> 261.666667
## H.npnct17.log 3.039622e-02 <NA> 96.104478
## S.npnct23.log 2.760321e-02 A.npnct23.log 6531.000000
## S.npnct25.log 2.760321e-02 <NA> 6531.000000
## H.has.ebola 2.588140e-02 <NA> 73.227273
## H.npnct01.log 2.271577e-02 <NA> 282.913043
## PubDate.month.fctr 1.914874e-02 <NA> 1.017514
## A.T.said 1.839871e-02 S.T.said 190.242424
## S.T.said 1.826884e-02 <NA> 190.242424
## PubDate.POSIX 1.568326e-02 <NA> 1.000000
## PubDate.zoo 1.568326e-02 <NA> 1.000000
## A.npnct23.log 1.537569e-02 A.npnct25.log 3264.500000
## A.npnct25.log 1.537569e-02 <NA> 3264.500000
## H.T.make 1.430572e-02 <NA> 322.200000
## H.npnct12.log 1.333613e-02 <NA> 4.937442
## myCategory.fctr 1.234541e-02 <NA> 1.337185
## UniqueID 1.182492e-02 <NA> 1.000000
## A.T.one 1.081694e-02 S.T.one 240.000000
## S.T.one 1.080534e-02 <NA> 240.038462
## A.T.state 1.020706e-02 S.T.state 315.700000
## S.T.state 1.012205e-02 <NA> 315.750000
## H.npnct03.log 9.533020e-03 <NA> 2176.333333
## A.npnct26.log 9.890046e-19 <NA> 0.000000
## H.npnct26.log 9.890046e-19 <NA> 0.000000
## S.npnct26.log 9.890046e-19 <NA> 0.000000
## H.T.take 8.582583e-04 <NA> 306.904762
## A.npnct17.log 1.587454e-03 <NA> 434.133333
## S.npnct17.log 1.587454e-03 <NA> 434.133333
## A.T.presid 1.789086e-03 <NA> 241.692308
## S.T.presid 2.079562e-03 <NA> 241.692308
## S.npnct08.log 2.413868e-03 <NA> 175.513514
## H.T.time 2.527450e-03 <NA> 247.538462
## A.npnct08.log 3.258100e-03 <NA> 170.868421
## S.npnct09.log 3.986882e-03 <NA> 175.486486
## A.npnct09.log 4.775988e-03 <NA> 170.842105
## A.npnct27.log 5.547032e-03 <NA> 6531.000000
## A.npnct11.log 5.547032e-03 <NA> 6531.000000
## H.npnct11.log 5.547032e-03 <NA> 6531.000000
## H.npnct22.log 5.547032e-03 <NA> 6531.000000
## S.npnct02.log 5.547032e-03 <NA> 6531.000000
## S.npnct11.log 5.547032e-03 <NA> 6531.000000
## PubDate.last100.log 7.663322e-03 <NA> 25.000000
## .rnorm 8.244230e-03 <NA> 2.000000
## H.npnct05.log 9.653967e-03 <NA> 543.333333
## H.T.obama 9.907543e-03 <NA> 229.750000
## H.T.say 9.960773e-03 <NA> 247.461538
## H.T.bank 9.989139e-03 <NA> 221.689655
## PubDate.date.fctr 1.164756e-02 <NA> 1.021394
## PubDate.second.fctr 1.187946e-02 <NA> 1.018204
## H.npnct07.log 1.201741e-02 <NA> 5.437234
## A.npnct07.log 1.214357e-02 S.npnct07.log 1631.750000
## S.npnct07.log 1.214357e-02 <NA> 1631.750000
## S.npnct03.log 1.240734e-02 <NA> 1305.400000
## A.npnct19.log 1.271661e-02 <NA> 1631.500000
## H.npnct13.log 1.305305e-02 <NA> 13.126638
## A.has.http 1.359260e-02 A.npnct19.log 1087.666667
## A.npnct03.log 1.359260e-02 S.npnct03.log 1087.666667
## H.T.big 1.390748e-02 <NA> 403.562500
## A.npnct02.log 1.451467e-02 A.npnct18.log 1087.500000
## A.npnct18.log 1.451467e-02 A.npnct20.log 1087.500000
## A.npnct20.log 1.451467e-02 <NA> 1087.500000
## A.has.year.colon 1.755336e-02 S.has.year.colon 652.200000
## S.has.year.colon 1.755336e-02 <NA> 652.200000
## A.npnct22.log 1.923169e-02 S.npnct22.log 543.333333
## S.npnct22.log 1.923169e-02 <NA> 543.333333
## H.npnct02.log 2.001851e-02 <NA> 501.461538
## H.T.test 2.065255e-02 <NA> 280.000000
## S.npnct15.log 2.121844e-02 <NA> 203.062500
## S.T.take 2.275732e-02 <NA> 287.090909
## A.T.take 2.282555e-02 S.T.take 287.045455
## A.npnct06.log 2.389145e-02 S.npnct06.log 115.642857
## S.npnct06.log 2.389145e-02 <NA> 115.642857
## A.npnct15.log 2.407715e-02 A.npnct02.log 196.696970
## H.npnct14.log 2.524770e-02 <NA> 22.802326
## H.T.deal 2.559418e-02 <NA> 258.080000
## S.T.new 2.769558e-02 <NA> 107.872727
## A.T.new 2.782876e-02 S.T.new 107.836364
## H.T.billion 2.949817e-02 <NA> 229.892857
## H.T.polit 3.062866e-02 <NA> 126.254902
## H.T.china 3.144808e-02 <NA> 238.555556
## H.T.art 3.280483e-02 <NA> 307.333333
## PubDate.minute.fctr 3.407385e-02 <NA> 1.483365
## S.npnct13.log 3.638891e-02 <NA> 5.706263
## A.npnct13.log 3.760012e-02 S.npnct13.log 5.715368
## A.T.will 3.887937e-02 <NA> 114.711538
## S.T.will 3.892267e-02 A.T.will 114.750000
## PubDate.wkday.fctr 3.980129e-02 <NA> 1.003268
## H.T.pictur 3.993172e-02 <NA> 104.032258
## S.T.day 4.188671e-02 <NA> 89.600000
## S.T.show 4.193803e-02 <NA> 274.608696
## A.T.show 4.196129e-02 S.T.show 274.608696
## A.T.day 4.196599e-02 S.T.day 89.585714
## H.T.new 4.327803e-02 <NA> 116.333333
## S.npnct30.log 4.370037e-02 <NA> 134.791667
## A.npnct30.log 4.373349e-02 S.npnct30.log 126.862745
## H.T.news 4.415284e-02 <NA> 238.518519
## H.T.first 4.458885e-02 <NA> 194.727273
## H.T.X2014 4.497745e-02 <NA> 112.824561
## A.T.first 4.603341e-02 <NA> 203.709677
## S.T.first 4.617532e-02 A.T.first 203.709677
## A.T.year 4.721236e-02 <NA> 167.108108
## S.T.year 4.729011e-02 A.T.year 167.108108
## A.T.report 4.741555e-02 <NA> 78.362500
## S.T.report 4.746920e-02 A.T.report 78.362500
## A.T.compani 4.751471e-02 <NA> 140.227273
## S.T.compani 4.764341e-02 A.T.compani 140.227273
## H.T.morn 4.838380e-02 A.npnct30.log 165.205128
## H.T.busi 4.901905e-02 <NA> 229.428571
## A.npnct14.log 4.999563e-02 <NA> 4.603330
## A.T.share 5.070234e-02 S.T.share 218.448276
## S.T.share 5.070234e-02 <NA> 218.448276
## H.npnct04.log 5.126277e-02 H.T.billion 38.325301
## S.T.time 5.303654e-02 <NA> 65.096774
## A.T.time 5.313395e-02 S.T.time 65.086022
## S.npnct14.log 5.332519e-02 A.npnct14.log 4.672000
## A.T.articl 5.445243e-02 <NA> 85.500000
## S.T.articl 5.446201e-02 A.T.articl 85.500000
## H.T.newyork 5.650839e-02 <NA> 112.446429
## A.T.newyork 5.706083e-02 <NA> 88.724638
## S.T.newyork 5.712853e-02 A.T.newyork 88.724638
## H.T.today 5.831308e-02 H.T.polit 138.239130
## H.T.springsumm 5.943248e-02 <NA> 106.966667
## H.T.day 6.033488e-02 <NA> 86.547945
## H.npnct15.log 6.158577e-02 H.T.springsumm 52.983471
## H.T.report 6.244050e-02 <NA> 102.000000
## A.npnct04.log 6.294642e-02 S.npnct04.log 28.536364
## S.npnct04.log 6.294642e-02 <NA> 28.536364
## H.T.daili 6.299948e-02 H.T.report 102.903226
## H.T.X2015 6.601141e-02 H.npnct15.log 96.833333
## S.npnct16.log 6.770952e-02 <NA> 13.647191
## H.T.week 6.812724e-02 <NA> 64.071429
## A.npnct16.log 6.893301e-02 S.npnct16.log 13.482222
## A.T.intern 6.949870e-02 <NA> 137.347826
## S.T.intern 6.953750e-02 A.T.intern 137.347826
## H.has.year.colon 7.842875e-02 S.T.intern 32.670103
## H.T.fashion 7.947505e-02 <NA> 76.926829
## H.npnct16.log 8.273237e-02 <NA> 3.914910
## A.T.fashion 8.419345e-02 <NA> 59.809524
## S.T.fashion 8.419711e-02 H.T.X2015 59.809524
## A.T.week 8.492895e-02 <NA> 57.122642
## S.T.week 8.503373e-02 A.T.week 57.122642
## H.npnct30.log 8.917338e-02 <NA> 24.123077
## S.npnct12.log 9.158156e-02 <NA> 1.660473
## A.npnct12.log 9.183870e-02 S.npnct12.log 1.660473
## H.ndgts.log 1.196633e-01 <NA> 13.616137
## S.ndgts.log 1.242046e-01 <NA> 10.511247
## A.ndgts.log 1.249484e-01 S.ndgts.log 10.501022
## H.nuppr.log 1.278085e-01 <NA> 1.033930
## H.nchrs.log 1.710624e-01 <NA> 1.023810
## H.nwrds.unq.log 2.044964e-01 H.nuppr.log 1.019071
## A.nchrs.log 2.245488e-01 <NA> 1.328571
## S.nchrs.log 2.246930e-01 A.nchrs.log 1.328571
## A.nwrds.unq.log 2.506012e-01 <NA> 1.061567
## S.nwrds.unq.log 2.507969e-01 S.nchrs.log 1.061567
## S.nuppr.log 2.718459e-01 <NA> 1.152620
## A.nuppr.log 2.720962e-01 S.nuppr.log 1.151308
## A.npnct05.log NA <NA> 0.000000
## A.npnct10.log NA <NA> 0.000000
## A.npnct24.log NA <NA> 0.000000
## A.npnct28.log NA <NA> 0.000000
## A.npnct29.log NA <NA> 0.000000
## A.npnct31.log NA <NA> 0.000000
## A.npnct32.log NA <NA> 0.000000
## clusterid NA <NA> 0.000000
## H.has.http NA <NA> 0.000000
## H.npnct10.log NA <NA> 0.000000
## H.npnct18.log NA <NA> 0.000000
## H.npnct19.log NA <NA> 0.000000
## H.npnct20.log NA <NA> 0.000000
## H.npnct23.log NA <NA> 0.000000
## H.npnct24.log NA <NA> 0.000000
## H.npnct25.log NA <NA> 0.000000
## H.npnct27.log NA <NA> 0.000000
## H.npnct28.log NA <NA> 0.000000
## H.npnct29.log NA <NA> 0.000000
## H.npnct31.log NA <NA> 0.000000
## H.npnct32.log NA <NA> 0.000000
## PubDate.year.fctr NA <NA> 0.000000
## S.has.http NA <NA> 0.000000
## S.npnct05.log NA <NA> 0.000000
## S.npnct10.log NA <NA> 0.000000
## S.npnct18.log NA <NA> 0.000000
## S.npnct19.log NA <NA> 0.000000
## S.npnct20.log NA <NA> 0.000000
## S.npnct24.log NA <NA> 0.000000
## S.npnct27.log NA <NA> 0.000000
## S.npnct28.log NA <NA> 0.000000
## S.npnct29.log NA <NA> 0.000000
## S.npnct31.log NA <NA> 0.000000
## S.npnct32.log NA <NA> 0.000000
## percentUnique zeroVar nzv myNearZV is.cor.y.abs.low
## Popular 0.03061849 FALSE FALSE FALSE FALSE
## WordCount.log 24.14268218 FALSE FALSE FALSE FALSE
## WordCount 24.15799143 FALSE FALSE FALSE FALSE
## H.nwrds.log 84.12431108 FALSE FALSE FALSE FALSE
## S.nwrds.log 93.67728108 FALSE FALSE FALSE FALSE
## PubDate.hour.fctr 0.04592774 FALSE FALSE FALSE FALSE
## A.nwrds.log 93.35578690 FALSE FALSE FALSE FALSE
## H.npnct21.log 0.06123699 FALSE FALSE FALSE FALSE
## PubDate.wkend 0.03061849 FALSE FALSE FALSE FALSE
## S.npnct21.log 0.07654623 FALSE FALSE FALSE FALSE
## A.npnct21.log 0.07654623 FALSE FALSE FALSE FALSE
## PubDate.last10 79.05695040 FALSE FALSE FALSE FALSE
## H.npnct08.log 0.03061849 FALSE TRUE FALSE FALSE
## H.npnct09.log 0.03061849 FALSE TRUE FALSE FALSE
## PubDate.last10.log 79.05695040 FALSE FALSE FALSE FALSE
## PubDate.last1.log 36.49724434 FALSE FALSE FALSE FALSE
## PubDate.last100 92.52908757 FALSE FALSE FALSE FALSE
## A.T.make 0.48989590 FALSE TRUE FALSE FALSE
## S.T.make 0.47458665 FALSE TRUE FALSE FALSE
## PubDate.last1 36.49724434 FALSE FALSE FALSE FALSE
## H.npnct06.log 0.06123699 FALSE TRUE FALSE FALSE
## A.T.can 0.48989590 FALSE TRUE FALSE FALSE
## A.npnct01.log 0.06123699 FALSE TRUE FALSE FALSE
## S.npnct01.log 0.06123699 FALSE TRUE FALSE FALSE
## S.T.can 0.41334966 FALSE TRUE FALSE FALSE
## H.npnct17.log 0.06123699 FALSE TRUE FALSE FALSE
## S.npnct23.log 0.03061849 FALSE TRUE TRUE FALSE
## S.npnct25.log 0.03061849 FALSE TRUE TRUE FALSE
## H.has.ebola 0.03061849 FALSE TRUE FALSE FALSE
## H.npnct01.log 0.04592774 FALSE TRUE FALSE FALSE
## PubDate.month.fctr 0.04592774 FALSE FALSE FALSE FALSE
## A.T.said 0.39804042 FALSE TRUE FALSE FALSE
## S.T.said 0.36742192 FALSE TRUE FALSE FALSE
## PubDate.POSIX 99.86221678 FALSE FALSE FALSE FALSE
## PubDate.zoo 99.86221678 FALSE FALSE FALSE FALSE
## A.npnct23.log 0.04592774 FALSE TRUE TRUE FALSE
## A.npnct25.log 0.04592774 FALSE TRUE TRUE FALSE
## H.T.make 0.13778322 FALSE TRUE FALSE FALSE
## H.npnct12.log 0.07654623 FALSE FALSE FALSE FALSE
## myCategory.fctr 0.30618494 FALSE FALSE FALSE FALSE
## UniqueID 100.00000000 FALSE FALSE FALSE FALSE
## A.T.one 0.50520514 FALSE TRUE FALSE FALSE
## S.T.one 0.45927740 FALSE TRUE FALSE FALSE
## A.T.state 0.42865891 FALSE TRUE FALSE FALSE
## S.T.state 0.42865891 FALSE TRUE FALSE FALSE
## H.npnct03.log 0.03061849 FALSE TRUE TRUE FALSE
## A.npnct26.log 0.01530925 TRUE TRUE TRUE TRUE
## H.npnct26.log 0.01530925 TRUE TRUE TRUE TRUE
## S.npnct26.log 0.01530925 TRUE TRUE TRUE TRUE
## H.T.take 0.15309247 FALSE TRUE FALSE TRUE
## A.npnct17.log 0.04592774 FALSE TRUE FALSE TRUE
## S.npnct17.log 0.04592774 FALSE TRUE FALSE TRUE
## A.T.presid 0.45927740 FALSE TRUE FALSE TRUE
## S.T.presid 0.42865891 FALSE TRUE FALSE TRUE
## S.npnct08.log 0.04592774 FALSE TRUE FALSE TRUE
## H.T.time 0.16840171 FALSE TRUE FALSE TRUE
## A.npnct08.log 0.04592774 FALSE TRUE FALSE TRUE
## S.npnct09.log 0.06123699 FALSE TRUE FALSE TRUE
## A.npnct09.log 0.06123699 FALSE TRUE FALSE TRUE
## A.npnct27.log 0.03061849 FALSE TRUE TRUE TRUE
## A.npnct11.log 0.03061849 FALSE TRUE TRUE TRUE
## H.npnct11.log 0.03061849 FALSE TRUE TRUE TRUE
## H.npnct22.log 0.03061849 FALSE TRUE TRUE TRUE
## S.npnct02.log 0.03061849 FALSE TRUE TRUE TRUE
## S.npnct11.log 0.03061849 FALSE TRUE TRUE TRUE
## PubDate.last100.log 92.19228414 FALSE FALSE FALSE TRUE
## .rnorm 99.98469075 FALSE FALSE FALSE FALSE
## H.npnct05.log 0.03061849 FALSE TRUE FALSE FALSE
## H.T.obama 0.16840171 FALSE TRUE FALSE FALSE
## H.T.say 0.16840171 FALSE TRUE FALSE FALSE
## H.T.bank 0.13778322 FALSE TRUE FALSE FALSE
## PubDate.date.fctr 0.07654623 FALSE FALSE FALSE FALSE
## PubDate.second.fctr 0.06123699 FALSE FALSE FALSE FALSE
## H.npnct07.log 0.12247397 FALSE FALSE FALSE FALSE
## A.npnct07.log 0.04592774 FALSE TRUE FALSE FALSE
## S.npnct07.log 0.04592774 FALSE TRUE FALSE FALSE
## S.npnct03.log 0.03061849 FALSE TRUE FALSE FALSE
## A.npnct19.log 0.06123699 FALSE TRUE FALSE FALSE
## H.npnct13.log 0.09185548 FALSE FALSE FALSE FALSE
## A.has.http 0.03061849 FALSE TRUE FALSE FALSE
## A.npnct03.log 0.03061849 FALSE TRUE FALSE FALSE
## H.T.big 0.19902021 FALSE TRUE FALSE FALSE
## A.npnct02.log 0.04592774 FALSE TRUE FALSE FALSE
## A.npnct18.log 0.04592774 FALSE TRUE FALSE FALSE
## A.npnct20.log 0.04592774 FALSE TRUE FALSE FALSE
## A.has.year.colon 0.03061849 FALSE TRUE FALSE FALSE
## S.has.year.colon 0.03061849 FALSE TRUE FALSE FALSE
## A.npnct22.log 0.03061849 FALSE TRUE FALSE FALSE
## S.npnct22.log 0.03061849 FALSE TRUE FALSE FALSE
## H.npnct02.log 0.03061849 FALSE TRUE FALSE FALSE
## H.T.test 0.13778322 FALSE TRUE FALSE FALSE
## S.npnct15.log 0.04592774 FALSE TRUE FALSE FALSE
## S.T.take 0.38273117 FALSE TRUE FALSE FALSE
## A.T.take 0.42865891 FALSE TRUE FALSE FALSE
## A.npnct06.log 0.03061849 FALSE TRUE FALSE FALSE
## S.npnct06.log 0.03061849 FALSE TRUE FALSE FALSE
## A.npnct15.log 0.10716473 FALSE TRUE FALSE FALSE
## H.npnct14.log 0.12247397 FALSE TRUE FALSE FALSE
## H.T.deal 0.13778322 FALSE TRUE FALSE FALSE
## S.T.new 0.48989590 FALSE TRUE FALSE FALSE
## A.T.new 0.50520514 FALSE TRUE FALSE FALSE
## H.T.billion 0.13778322 FALSE TRUE FALSE FALSE
## H.T.polit 0.13778322 FALSE TRUE FALSE FALSE
## H.T.china 0.18371096 FALSE TRUE FALSE FALSE
## H.T.art 0.19902021 FALSE TRUE FALSE FALSE
## PubDate.minute.fctr 0.06123699 FALSE FALSE FALSE FALSE
## S.npnct13.log 0.09185548 FALSE FALSE FALSE FALSE
## A.npnct13.log 0.12247397 FALSE FALSE FALSE FALSE
## A.T.will 0.62767912 FALSE TRUE FALSE FALSE
## S.T.will 0.55113288 FALSE TRUE FALSE FALSE
## PubDate.wkday.fctr 0.10716473 FALSE FALSE FALSE FALSE
## H.T.pictur 0.10716473 FALSE TRUE FALSE FALSE
## S.T.day 0.39804042 FALSE TRUE FALSE FALSE
## S.T.show 0.39804042 FALSE TRUE FALSE FALSE
## A.T.show 0.41334966 FALSE TRUE FALSE FALSE
## A.T.day 0.42865891 FALSE TRUE FALSE FALSE
## H.T.new 0.19902021 FALSE TRUE FALSE FALSE
## S.npnct30.log 0.04592774 FALSE TRUE FALSE FALSE
## A.npnct30.log 0.04592774 FALSE TRUE FALSE FALSE
## H.T.news 0.16840171 FALSE TRUE FALSE FALSE
## H.T.first 0.15309247 FALSE TRUE FALSE FALSE
## H.T.X2014 0.13778322 FALSE TRUE FALSE FALSE
## A.T.first 0.42865891 FALSE TRUE FALSE FALSE
## S.T.first 0.39804042 FALSE TRUE FALSE FALSE
## A.T.year 0.44396816 FALSE TRUE FALSE FALSE
## S.T.year 0.42865891 FALSE TRUE FALSE FALSE
## A.T.report 0.38273117 FALSE TRUE FALSE FALSE
## S.T.report 0.35211268 FALSE TRUE FALSE FALSE
## A.T.compani 0.50520514 FALSE TRUE FALSE FALSE
## S.T.compani 0.44396816 FALSE TRUE FALSE FALSE
## H.T.morn 0.07654623 FALSE TRUE FALSE FALSE
## H.T.busi 0.18371096 FALSE TRUE FALSE FALSE
## A.npnct14.log 0.16840171 FALSE FALSE FALSE FALSE
## A.T.share 0.36742192 FALSE TRUE FALSE FALSE
## S.T.share 0.36742192 FALSE TRUE FALSE FALSE
## H.npnct04.log 0.04592774 FALSE TRUE FALSE FALSE
## S.T.time 0.47458665 FALSE TRUE FALSE FALSE
## A.T.time 0.47458665 FALSE TRUE FALSE FALSE
## S.npnct14.log 0.16840171 FALSE FALSE FALSE FALSE
## A.T.articl 0.27556644 FALSE TRUE FALSE FALSE
## S.T.articl 0.27556644 FALSE TRUE FALSE FALSE
## H.T.newyork 0.15309247 FALSE TRUE FALSE FALSE
## A.T.newyork 0.42865891 FALSE TRUE FALSE FALSE
## S.T.newyork 0.41334966 FALSE TRUE FALSE FALSE
## H.T.today 0.13778322 FALSE TRUE FALSE FALSE
## H.T.springsumm 0.09185548 FALSE TRUE FALSE FALSE
## H.T.day 0.18371096 FALSE TRUE FALSE FALSE
## H.npnct15.log 0.03061849 FALSE TRUE FALSE FALSE
## H.T.report 0.16840171 FALSE TRUE FALSE FALSE
## A.npnct04.log 0.07654623 FALSE TRUE FALSE FALSE
## S.npnct04.log 0.07654623 FALSE TRUE FALSE FALSE
## H.T.daili 0.16840171 FALSE TRUE FALSE FALSE
## H.T.X2015 0.10716473 FALSE TRUE FALSE FALSE
## S.npnct16.log 0.04592774 FALSE FALSE FALSE FALSE
## H.T.week 0.16840171 FALSE TRUE FALSE FALSE
## A.npnct16.log 0.04592774 FALSE FALSE FALSE FALSE
## A.T.intern 0.32149418 FALSE TRUE FALSE FALSE
## S.T.intern 0.30618494 FALSE TRUE FALSE FALSE
## H.has.year.colon 0.03061849 FALSE TRUE FALSE FALSE
## H.T.fashion 0.19902021 FALSE TRUE FALSE FALSE
## H.npnct16.log 0.04592774 FALSE FALSE FALSE FALSE
## A.T.fashion 0.41334966 FALSE TRUE FALSE FALSE
## S.T.fashion 0.39804042 FALSE TRUE FALSE FALSE
## A.T.week 0.48989590 FALSE TRUE FALSE FALSE
## S.T.week 0.42865891 FALSE TRUE FALSE FALSE
## H.npnct30.log 0.03061849 FALSE TRUE FALSE FALSE
## S.npnct12.log 0.13778322 FALSE FALSE FALSE FALSE
## A.npnct12.log 0.13778322 FALSE FALSE FALSE FALSE
## H.ndgts.log 0.18371096 FALSE FALSE FALSE FALSE
## S.ndgts.log 0.26025720 FALSE FALSE FALSE FALSE
## A.ndgts.log 0.29087569 FALSE FALSE FALSE FALSE
## H.nuppr.log 0.29087569 FALSE FALSE FALSE FALSE
## H.nchrs.log 1.57685242 FALSE FALSE FALSE FALSE
## H.nwrds.unq.log 0.21432945 FALSE FALSE FALSE FALSE
## A.nchrs.log 4.39375383 FALSE FALSE FALSE FALSE
## S.nchrs.log 3.72014697 FALSE FALSE FALSE FALSE
## A.nwrds.unq.log 0.55113288 FALSE FALSE FALSE FALSE
## S.nwrds.unq.log 0.44396816 FALSE FALSE FALSE FALSE
## S.nuppr.log 0.33680343 FALSE FALSE FALSE FALSE
## A.nuppr.log 0.33680343 FALSE FALSE FALSE FALSE
## A.npnct05.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct10.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct24.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct28.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct29.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct31.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct32.log 0.01530925 TRUE TRUE TRUE NA
## clusterid 0.01530925 TRUE TRUE TRUE NA
## H.has.http 0.01530925 TRUE TRUE TRUE NA
## H.npnct10.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct18.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct19.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct20.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct23.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct24.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct25.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct27.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct28.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct29.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct31.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct32.log 0.01530925 TRUE TRUE TRUE NA
## PubDate.year.fctr 0.01530925 TRUE TRUE TRUE NA
## S.has.http 0.01530925 TRUE TRUE TRUE NA
## S.npnct05.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct10.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct18.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct19.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct20.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct24.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct27.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct28.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct29.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct31.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct32.log 0.01530925 TRUE TRUE TRUE NAprint(myplot_scatter(glb_feats_df, "percentUnique", "freqRatio",
colorcol_name="myNearZV", jitter=TRUE) +
geom_point(aes(shape=nzv)) + xlim(-5, 25))## Warning in myplot_scatter(glb_feats_df, "percentUnique", "freqRatio",
## colorcol_name = "myNearZV", : converting myNearZV to class:factor## Warning: Removed 13 rows containing missing values (geom_point).## Warning: Removed 13 rows containing missing values (geom_point).## Warning: Removed 13 rows containing missing values (geom_point).
print(subset(glb_feats_df, myNearZV))## id cor.y exclude.as.feat
## S.npnct23.log S.npnct23.log 2.760321e-02 0
## S.npnct25.log S.npnct25.log 2.760321e-02 0
## A.npnct23.log A.npnct23.log 1.537569e-02 0
## A.npnct25.log A.npnct25.log 1.537569e-02 0
## H.npnct03.log H.npnct03.log 9.533020e-03 0
## A.npnct26.log A.npnct26.log -9.890046e-19 0
## H.npnct26.log H.npnct26.log -9.890046e-19 0
## S.npnct26.log S.npnct26.log -9.890046e-19 0
## A.npnct27.log A.npnct27.log -5.547032e-03 0
## A.npnct11.log A.npnct11.log -5.547032e-03 0
## H.npnct11.log H.npnct11.log -5.547032e-03 0
## H.npnct22.log H.npnct22.log -5.547032e-03 0
## S.npnct02.log S.npnct02.log -5.547032e-03 0
## S.npnct11.log S.npnct11.log -5.547032e-03 0
## A.npnct05.log A.npnct05.log NA 0
## A.npnct10.log A.npnct10.log NA 0
## A.npnct24.log A.npnct24.log NA 0
## A.npnct28.log A.npnct28.log NA 0
## A.npnct29.log A.npnct29.log NA 0
## A.npnct31.log A.npnct31.log NA 0
## A.npnct32.log A.npnct32.log NA 0
## clusterid clusterid NA 0
## H.has.http H.has.http NA 0
## H.npnct10.log H.npnct10.log NA 0
## H.npnct18.log H.npnct18.log NA 0
## H.npnct19.log H.npnct19.log NA 0
## H.npnct20.log H.npnct20.log NA 0
## H.npnct23.log H.npnct23.log NA 0
## H.npnct24.log H.npnct24.log NA 0
## H.npnct25.log H.npnct25.log NA 0
## H.npnct27.log H.npnct27.log NA 0
## H.npnct28.log H.npnct28.log NA 0
## H.npnct29.log H.npnct29.log NA 0
## H.npnct31.log H.npnct31.log NA 0
## H.npnct32.log H.npnct32.log NA 0
## PubDate.year.fctr PubDate.year.fctr NA 0
## S.has.http S.has.http NA 0
## S.npnct05.log S.npnct05.log NA 0
## S.npnct10.log S.npnct10.log NA 0
## S.npnct18.log S.npnct18.log NA 0
## S.npnct19.log S.npnct19.log NA 0
## S.npnct20.log S.npnct20.log NA 0
## S.npnct24.log S.npnct24.log NA 0
## S.npnct27.log S.npnct27.log NA 0
## S.npnct28.log S.npnct28.log NA 0
## S.npnct29.log S.npnct29.log NA 0
## S.npnct31.log S.npnct31.log NA 0
## S.npnct32.log S.npnct32.log NA 0
## cor.y.abs cor.high.X freqRatio percentUnique
## S.npnct23.log 2.760321e-02 A.npnct23.log 6531.000 0.03061849
## S.npnct25.log 2.760321e-02 <NA> 6531.000 0.03061849
## A.npnct23.log 1.537569e-02 A.npnct25.log 3264.500 0.04592774
## A.npnct25.log 1.537569e-02 <NA> 3264.500 0.04592774
## H.npnct03.log 9.533020e-03 <NA> 2176.333 0.03061849
## A.npnct26.log 9.890046e-19 <NA> 0.000 0.01530925
## H.npnct26.log 9.890046e-19 <NA> 0.000 0.01530925
## S.npnct26.log 9.890046e-19 <NA> 0.000 0.01530925
## A.npnct27.log 5.547032e-03 <NA> 6531.000 0.03061849
## A.npnct11.log 5.547032e-03 <NA> 6531.000 0.03061849
## H.npnct11.log 5.547032e-03 <NA> 6531.000 0.03061849
## H.npnct22.log 5.547032e-03 <NA> 6531.000 0.03061849
## S.npnct02.log 5.547032e-03 <NA> 6531.000 0.03061849
## S.npnct11.log 5.547032e-03 <NA> 6531.000 0.03061849
## A.npnct05.log NA <NA> 0.000 0.01530925
## A.npnct10.log NA <NA> 0.000 0.01530925
## A.npnct24.log NA <NA> 0.000 0.01530925
## A.npnct28.log NA <NA> 0.000 0.01530925
## A.npnct29.log NA <NA> 0.000 0.01530925
## A.npnct31.log NA <NA> 0.000 0.01530925
## A.npnct32.log NA <NA> 0.000 0.01530925
## clusterid NA <NA> 0.000 0.01530925
## H.has.http NA <NA> 0.000 0.01530925
## H.npnct10.log NA <NA> 0.000 0.01530925
## H.npnct18.log NA <NA> 0.000 0.01530925
## H.npnct19.log NA <NA> 0.000 0.01530925
## H.npnct20.log NA <NA> 0.000 0.01530925
## H.npnct23.log NA <NA> 0.000 0.01530925
## H.npnct24.log NA <NA> 0.000 0.01530925
## H.npnct25.log NA <NA> 0.000 0.01530925
## H.npnct27.log NA <NA> 0.000 0.01530925
## H.npnct28.log NA <NA> 0.000 0.01530925
## H.npnct29.log NA <NA> 0.000 0.01530925
## H.npnct31.log NA <NA> 0.000 0.01530925
## H.npnct32.log NA <NA> 0.000 0.01530925
## PubDate.year.fctr NA <NA> 0.000 0.01530925
## S.has.http NA <NA> 0.000 0.01530925
## S.npnct05.log NA <NA> 0.000 0.01530925
## S.npnct10.log NA <NA> 0.000 0.01530925
## S.npnct18.log NA <NA> 0.000 0.01530925
## S.npnct19.log NA <NA> 0.000 0.01530925
## S.npnct20.log NA <NA> 0.000 0.01530925
## S.npnct24.log NA <NA> 0.000 0.01530925
## S.npnct27.log NA <NA> 0.000 0.01530925
## S.npnct28.log NA <NA> 0.000 0.01530925
## S.npnct29.log NA <NA> 0.000 0.01530925
## S.npnct31.log NA <NA> 0.000 0.01530925
## S.npnct32.log NA <NA> 0.000 0.01530925
## zeroVar nzv myNearZV is.cor.y.abs.low
## S.npnct23.log FALSE TRUE TRUE FALSE
## S.npnct25.log FALSE TRUE TRUE FALSE
## A.npnct23.log FALSE TRUE TRUE FALSE
## A.npnct25.log FALSE TRUE TRUE FALSE
## H.npnct03.log FALSE TRUE TRUE FALSE
## A.npnct26.log TRUE TRUE TRUE TRUE
## H.npnct26.log TRUE TRUE TRUE TRUE
## S.npnct26.log TRUE TRUE TRUE TRUE
## A.npnct27.log FALSE TRUE TRUE TRUE
## A.npnct11.log FALSE TRUE TRUE TRUE
## H.npnct11.log FALSE TRUE TRUE TRUE
## H.npnct22.log FALSE TRUE TRUE TRUE
## S.npnct02.log FALSE TRUE TRUE TRUE
## S.npnct11.log FALSE TRUE TRUE TRUE
## A.npnct05.log TRUE TRUE TRUE NA
## A.npnct10.log TRUE TRUE TRUE NA
## A.npnct24.log TRUE TRUE TRUE NA
## A.npnct28.log TRUE TRUE TRUE NA
## A.npnct29.log TRUE TRUE TRUE NA
## A.npnct31.log TRUE TRUE TRUE NA
## A.npnct32.log TRUE TRUE TRUE NA
## clusterid TRUE TRUE TRUE NA
## H.has.http TRUE TRUE TRUE NA
## H.npnct10.log TRUE TRUE TRUE NA
## H.npnct18.log TRUE TRUE TRUE NA
## H.npnct19.log TRUE TRUE TRUE NA
## H.npnct20.log TRUE TRUE TRUE NA
## H.npnct23.log TRUE TRUE TRUE NA
## H.npnct24.log TRUE TRUE TRUE NA
## H.npnct25.log TRUE TRUE TRUE NA
## H.npnct27.log TRUE TRUE TRUE NA
## H.npnct28.log TRUE TRUE TRUE NA
## H.npnct29.log TRUE TRUE TRUE NA
## H.npnct31.log TRUE TRUE TRUE NA
## H.npnct32.log TRUE TRUE TRUE NA
## PubDate.year.fctr TRUE TRUE TRUE NA
## S.has.http TRUE TRUE TRUE NA
## S.npnct05.log TRUE TRUE TRUE NA
## S.npnct10.log TRUE TRUE TRUE NA
## S.npnct18.log TRUE TRUE TRUE NA
## S.npnct19.log TRUE TRUE TRUE NA
## S.npnct20.log TRUE TRUE TRUE NA
## S.npnct24.log TRUE TRUE TRUE NA
## S.npnct27.log TRUE TRUE TRUE NA
## S.npnct28.log TRUE TRUE TRUE NA
## S.npnct29.log TRUE TRUE TRUE NA
## S.npnct31.log TRUE TRUE TRUE NA
## S.npnct32.log TRUE TRUE TRUE NAglb_entity_df <- glb_entity_df[, setdiff(names(glb_entity_df),
subset(glb_feats_df, myNearZV)$id)]
glb_chunks_df <- myadd_chunk(glb_chunks_df, "partition.data.training", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 8 select.features 5 0 182.126 234.906 52.78
## 9 partition.data.training 6 0 234.906 NA NAStep 6.0: partition data training
if (all(is.na(glb_newent_df[, glb_rsp_var]))) {
require(caTools)
set.seed(glb_split_sample.seed)
split <- sample.split(glb_trnent_df[, glb_rsp_var_raw],
SplitRatio=1 - (nrow(glb_newent_df) * 1.1 / nrow(glb_trnent_df)))
glb_fitent_df <- glb_trnent_df[split, ]
glb_OOBent_df <- glb_trnent_df[!split ,]
} else {
print(sprintf("Newdata contains non-NA data for %s; setting OOB to Newdata",
glb_rsp_var))
glb_fitent_df <- glb_trnent_df; glb_OOBent_df <- glb_newent_df
}## Loading required package: caToolsif (!is.null(glb_max_fitent_obs) && (nrow(glb_fitent_df) > glb_max_fitent_obs)) {
warning("glb_fitent_df restricted to glb_max_fitent_obs: ",
format(glb_max_fitent_obs, big.mark=","))
org_fitent_df <- glb_fitent_df
glb_fitent_df <-
org_fitent_df[split <- sample.split(org_fitent_df[, glb_rsp_var_raw],
SplitRatio=glb_max_fitent_obs), ]
org_fitent_df <- NULL
}
sav_entity_df <- glb_entity_df
glb_entity_df$.lcn <- ""
glb_entity_df[glb_entity_df[, glb_id_vars] %in%
glb_fitent_df[, glb_id_vars], ".lcn"] <- "Fit"
glb_entity_df[glb_entity_df[, glb_id_vars] %in%
glb_OOBent_df[, glb_id_vars], ".lcn"] <- "OOB"
dsp_class_dstrb <- function(obs_df, location_var, partition_var) {
xtab_df <- mycreate_xtab_df(obs_df, c(location_var, partition_var))
rownames(xtab_df) <- xtab_df[, location_var]
xtab_df <- xtab_df[, -grepl(location_var, names(xtab_df))]
print(xtab_df)
print(xtab_df / rowSums(xtab_df, na.rm=TRUE))
}
# Ensure proper splits by glb_rsp_var_raw & user-specified feature for OOB vs. new
dsp_class_dstrb(glb_entity_df, ".lcn", glb_rsp_var_raw)## Popular.0 Popular.1 Popular.NA
## NA NA 1870
## Fit 3726 749 NA
## OOB 1713 344 NA
## Popular.0 Popular.1 Popular.NA
## NA NA 1
## Fit 0.8326257 0.1673743 NA
## OOB 0.8327662 0.1672338 NAnewent_ctgry_df <- mycreate_sqlxtab_df(subset(glb_entity_df, .src == "Test"),
"myCategory")
OOBent_ctgry_df <- mycreate_sqlxtab_df(subset(glb_entity_df, .lcn == "OOB"),
"myCategory")
glb_ctgry_df <- merge(newent_ctgry_df, OOBent_ctgry_df, by="myCategory", all=TRUE,
suffixes=c(".Tst", ".OOB"))
glb_ctgry_df$.freqRatio.Tst <- glb_ctgry_df$.n.Tst / sum(glb_ctgry_df$.n.Tst, na.rm=TRUE)
glb_ctgry_df$.freqRatio.OOB <- glb_ctgry_df$.n.OOB / sum(glb_ctgry_df$.n.OOB, na.rm=TRUE)
print(orderBy(~-.freqRatio.Tst-.freqRatio.OOB, glb_ctgry_df))## myCategory .n.Tst .n.OOB .freqRatio.Tst
## 1 ## 338 407 0.180748663
## 6 Business#Business Day#Dealbook 304 312 0.162566845
## 10 Culture#Arts# 244 225 0.130481283
## 15 OpEd#Opinion# 164 154 0.087700535
## 9 Business#Technology# 113 114 0.060427807
## 20 TStyle## 105 221 0.056149733
## 5 #U.S.#Education 90 93 0.048128342
## 13 Metro#N.Y. / Region# 67 60 0.035828877
## 18 Styles#U.S.# 62 54 0.033155080
## 16 Science#Health# 57 66 0.030481283
## 12 Foreign#World#Asia Pacific 56 61 0.029946524
## 2 #Multimedia# 52 42 0.027807487
## 11 Foreign#World# 47 47 0.025133690
## 7 Business#Business Day#Small Business 42 45 0.022459893
## 8 Business#Crosswords/Games# 42 40 0.022459893
## 19 Travel#Travel# 35 31 0.018716578
## 3 #Opinion#Room For Debate 24 21 0.012834225
## 17 Styles##Fashion 15 41 0.008021390
## 4 #Opinion#The Public Editor 10 10 0.005347594
## 14 myOther 3 13 0.001604278
## .freqRatio.OOB
## 1 0.197860963
## 6 0.151677200
## 10 0.109382596
## 15 0.074866310
## 9 0.055420515
## 20 0.107438017
## 5 0.045211473
## 13 0.029168692
## 18 0.026251823
## 16 0.032085561
## 12 0.029654837
## 2 0.020418085
## 11 0.022848809
## 7 0.021876519
## 8 0.019445795
## 19 0.015070491
## 3 0.010209042
## 17 0.019931940
## 4 0.004861449
## 14 0.006319883# Run this line by line
print("glb_feats_df:"); print(dim(glb_feats_df))## [1] "glb_feats_df:"## [1] 214 11sav_feats_df <- glb_feats_df
glb_feats_df <- sav_feats_df
glb_feats_df[, "rsp_var_raw"] <- FALSE
glb_feats_df[glb_feats_df$id == glb_rsp_var_raw, "rsp_var_raw"] <- TRUE
glb_feats_df$exclude.as.feat <- (glb_feats_df$exclude.as.feat == 1)
if (!is.null(glb_id_vars) && glb_id_vars != ".rownames")
glb_feats_df[glb_feats_df$id %in% glb_id_vars, "id_var"] <- TRUE
add_feats_df <- data.frame(id=glb_rsp_var, exclude.as.feat=TRUE, rsp_var=TRUE)
row.names(add_feats_df) <- add_feats_df$id; print(add_feats_df)## id exclude.as.feat rsp_var
## Popular.fctr Popular.fctr TRUE TRUEglb_feats_df <- myrbind_df(glb_feats_df, add_feats_df)
print(subset(glb_feats_df, rsp_var_raw | rsp_var | id_var))## id cor.y exclude.as.feat cor.y.abs cor.high.X
## Popular Popular 1.00000000 TRUE 1.00000000 <NA>
## UniqueID UniqueID 0.01182492 TRUE 0.01182492 <NA>
## Popular.fctr Popular.fctr NA TRUE NA <NA>
## freqRatio percentUnique zeroVar nzv myNearZV
## Popular 4.976212 0.03061849 FALSE FALSE FALSE
## UniqueID 1.000000 100.00000000 FALSE FALSE FALSE
## Popular.fctr NA NA NA NA NA
## is.cor.y.abs.low rsp_var_raw id_var rsp_var
## Popular FALSE TRUE NA NA
## UniqueID FALSE FALSE TRUE NA
## Popular.fctr NA NA NA TRUEprint("glb_feats_df vs. glb_entity_df: "); ## [1] "glb_feats_df vs. glb_entity_df: "print(setdiff(glb_feats_df$id, names(glb_entity_df)))## [1] "S.npnct23.log" "S.npnct25.log" "A.npnct23.log"
## [4] "A.npnct25.log" "H.npnct03.log" "A.npnct26.log"
## [7] "H.npnct26.log" "S.npnct26.log" "A.npnct27.log"
## [10] "A.npnct11.log" "H.npnct11.log" "H.npnct22.log"
## [13] "S.npnct02.log" "S.npnct11.log" "A.npnct05.log"
## [16] "A.npnct10.log" "A.npnct24.log" "A.npnct28.log"
## [19] "A.npnct29.log" "A.npnct31.log" "A.npnct32.log"
## [22] "clusterid" "H.has.http" "H.npnct10.log"
## [25] "H.npnct18.log" "H.npnct19.log" "H.npnct20.log"
## [28] "H.npnct23.log" "H.npnct24.log" "H.npnct25.log"
## [31] "H.npnct27.log" "H.npnct28.log" "H.npnct29.log"
## [34] "H.npnct31.log" "H.npnct32.log" "PubDate.year.fctr"
## [37] "S.has.http" "S.npnct05.log" "S.npnct10.log"
## [40] "S.npnct18.log" "S.npnct19.log" "S.npnct20.log"
## [43] "S.npnct24.log" "S.npnct27.log" "S.npnct28.log"
## [46] "S.npnct29.log" "S.npnct31.log" "S.npnct32.log"print("glb_entity_df vs. glb_feats_df: "); ## [1] "glb_entity_df vs. glb_feats_df: "# Ensure these are only chr vars
print(setdiff(setdiff(names(glb_entity_df), glb_feats_df$id),
myfind_chr_cols_df(glb_entity_df)))## character(0)#print(setdiff(setdiff(names(glb_entity_df), glb_exclude_vars_as_features),
# glb_feats_df$id))
print("glb_entity_df: "); print(dim(glb_entity_df))## [1] "glb_entity_df: "## [1] 8402 177print("glb_trnent_df: "); print(dim(glb_trnent_df))## [1] "glb_trnent_df: "## [1] 6532 224print("glb_fitent_df: "); print(dim(glb_fitent_df))## [1] "glb_fitent_df: "## [1] 4475 224print("glb_OOBent_df: "); print(dim(glb_OOBent_df))## [1] "glb_OOBent_df: "## [1] 2057 224print("glb_newent_df: "); print(dim(glb_newent_df))## [1] "glb_newent_df: "## [1] 1870 224# sav_entity_df <- glb_entity_df
# glb_entity_df <- sav_entity_df
# # Does not handle NULL or length(glb_id_vars) > 1
# glb_entity_df$.src.trn <- 0
# glb_entity_df[glb_entity_df[, glb_id_vars] %in% glb_trnent_df[, glb_id_vars],
# ".src.trn"] <- 1
# glb_entity_df$.src.fit <- 0
# glb_entity_df[glb_entity_df[, glb_id_vars] %in% glb_fitent_df[, glb_id_vars],
# ".src.fit"] <- 1
# glb_entity_df$.src.OOB <- 0
# glb_entity_df[glb_entity_df[, glb_id_vars] %in% glb_OOBent_df[, glb_id_vars],
# ".src.OOB"] <- 1
# glb_entity_df$.src.new <- 0
# glb_entity_df[glb_entity_df[, glb_id_vars] %in% glb_newent_df[, glb_id_vars],
# ".src.new"] <- 1
# #print(unique(glb_entity_df[, ".src.trn"]))
# write_cols <- c(glb_feats_df$id,
# ".src.trn", ".src.fit", ".src.OOB", ".src.new")
# glb_entity_df <- glb_entity_df[, write_cols]
#
# tmp_feats_df <- glb_feats_df
# tmp_entity_df <- glb_entity_df
if (glb_save_envir)
save(glb_feats_df,
glb_entity_df, #glb_trnent_df, glb_fitent_df, glb_OOBent_df, glb_newent_df,
file=paste0(glb_out_pfx, "blddfs_dsk.RData"))
# load(paste0(glb_out_pfx, "blddfs_dsk.RData"))
# if (!all.equal(tmp_feats_df, glb_feats_df))
# stop("glb_feats_df r/w not working")
# if (!all.equal(tmp_entity_df, glb_entity_df))
# stop("glb_entity_df r/w not working")
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 9 partition.data.training 6 0 234.906 236.208 1.303
## 10 fit.models 7 0 236.209 NA NAStep 7.0: fit models
# load(paste0(glb_out_pfx, "dsk.RData"))
# keep_cols <- setdiff(names(glb_entity_df),
# grep("^.src", names(glb_entity_df), value=TRUE))
# glb_trnent_df <- glb_entity_df[glb_entity_df$.src.trn == 1, keep_cols]
# glb_fitent_df <- glb_entity_df[glb_entity_df$.src.fit == 1, keep_cols]
# glb_OOBent_df <- glb_entity_df[glb_entity_df$.src.OOB == 1, keep_cols]
# glb_newent_df <- glb_entity_df[glb_entity_df$.src.new == 1, keep_cols]
#
# glb_models_lst <- list(); glb_models_df <- data.frame()
#
if (glb_is_classification && glb_is_binomial &&
(length(unique(glb_fitent_df[, glb_rsp_var])) < 2))
stop("glb_fitent_df$", glb_rsp_var, ": contains less than 2 unique values: ",
paste0(unique(glb_fitent_df[, glb_rsp_var]), collapse=", "))
max_cor_y_x_var <- orderBy(~ -cor.y.abs,
subset(glb_feats_df, (exclude.as.feat == 0) & !is.cor.y.abs.low))[1, "id"]
if (!is.null(glb_Baseline_mdl_var)) {
if ((max_cor_y_x_var != glb_Baseline_mdl_var) &
(glb_feats_df[max_cor_y_x_var, "cor.y.abs"] >
glb_feats_df[glb_Baseline_mdl_var, "cor.y.abs"]))
stop(max_cor_y_x_var, " has a lower correlation with ", glb_rsp_var,
" than the Baseline var: ", glb_Baseline_mdl_var)
}
glb_model_type <- ifelse(glb_is_regression, "regression", "classification")
# Baseline
if (!is.null(glb_Baseline_mdl_var))
ret_lst <- myfit_mdl_fn(model_id="Baseline", model_method="mybaseln_classfr",
indep_vars_vctr=glb_Baseline_mdl_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df)
# Most Frequent Outcome "MFO" model: mean(y) for regression
# Not using caret's nullModel since model stats not avl
# Cannot use rpart for multinomial classification since it predicts non-MFO
ret_lst <- myfit_mdl(model_id="MFO",
model_method=ifelse(glb_is_regression, "lm", "myMFO_classfr"),
model_type=glb_model_type,
indep_vars_vctr=".rnorm",
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df)## [1] "fitting model: MFO.myMFO_classfr"
## [1] " indep_vars: .rnorm"
## Fitting parameter = none on full training set
## [1] "in MFO.Classifier$fit"
## [1] "unique.vals:"
## [1] N Y
## Levels: N Y
## [1] "unique.prob:"
## y
## N Y
## 0.8326257 0.1673743
## [1] "MFO.val:"
## [1] "N"
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 -none- numeric
## MFO.val 1 -none- character
## x.names 1 -none- character
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## [1] " calling mypredict_mdl for fit:"## Loading required package: ROCR
## Loading required package: gplots
##
## Attaching package: 'gplots'
##
## The following object is masked from 'package:stats':
##
## lowess## [1] "entr MFO.Classifier$predict"
## [1] "exit MFO.Classifier$predict"
## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.8326257 0.1673743
## 2 0.8326257 0.1673743
## 3 0.8326257 0.1673743
## 4 0.8326257 0.1673743
## 5 0.8326257 0.1673743
## 6 0.8326257 0.1673743
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.MFO.myMFO_classfr.N
## 1 N 3726
## 2 Y 749
## Prediction
## Reference N Y
## N 3726 0
## Y 749 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.326257e-01 0.000000e+00 8.213602e-01 8.434553e-01 8.326257e-01
## AccuracyPValue McnemarPValue
## 5.097571e-01 1.800616e-164
## [1] " calling mypredict_mdl for OOB:"
## [1] "entr MFO.Classifier$predict"
## [1] "exit MFO.Classifier$predict"
## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.8326257 0.1673743
## 2 0.8326257 0.1673743
## 3 0.8326257 0.1673743
## 4 0.8326257 0.1673743
## 5 0.8326257 0.1673743
## 6 0.8326257 0.1673743
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.MFO.myMFO_classfr.N
## 1 N 1713
## 2 Y 344
## Prediction
## Reference N Y
## N 1713 0
## Y 344 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.327662e-01 0.000000e+00 8.159247e-01 8.486533e-01 8.327662e-01
## AccuracyPValue McnemarPValue
## 5.143944e-01 2.337097e-76
## model_id model_method feats max.nTuningRuns
## 1 MFO.myMFO_classfr myMFO_classfr .rnorm 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.666 0.003 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.8326257
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8213602 0.8434553 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.8327662
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8159247 0.8486533 0if (glb_is_classification)
# "random" model - only for classification;
# none needed for regression since it is same as MFO
ret_lst <- myfit_mdl(model_id="Random", model_method="myrandom_classfr",
model_type=glb_model_type,
indep_vars_vctr=".rnorm",
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df)## [1] "fitting model: Random.myrandom_classfr"
## [1] " indep_vars: .rnorm"
## Fitting parameter = none on full training set
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 table numeric
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## [1] " calling mypredict_mdl for fit:"
## [1] "in Random.Classifier$prob"
## threshold f.score
## 1 0.0 0.2867534
## 2 0.1 0.2867534
## 3 0.2 0.1735751
## 4 0.3 0.1735751
## 5 0.4 0.1735751
## 6 0.5 0.1735751
## 7 0.6 0.1735751
## 8 0.7 0.1735751
## 9 0.8 0.1735751
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.1000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.Random.myrandom_classfr.Y
## 1 N 3726
## 2 Y 749
## Prediction
## Reference N Y
## N 0 3726
## Y 0 749
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.1673743 0.0000000 0.1565447 0.1786398 0.8326257
## AccuracyPValue McnemarPValue
## 1.0000000 0.0000000
## [1] " calling mypredict_mdl for OOB:"
## [1] "in Random.Classifier$prob"
## threshold f.score
## 1 0.0 0.2865473
## 2 0.1 0.2865473
## 3 0.2 0.1547278
## 4 0.3 0.1547278
## 5 0.4 0.1547278
## 6 0.5 0.1547278
## 7 0.6 0.1547278
## 8 0.7 0.1547278
## 9 0.8 0.1547278
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.1000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.Random.myrandom_classfr.Y
## 1 N 1713
## 2 Y 344
## Prediction
## Reference N Y
## N 0 1713
## Y 0 344
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.1672338 0.0000000 0.1513467 0.1840753 0.8327662
## AccuracyPValue McnemarPValue
## 1.0000000 0.0000000
## model_id model_method feats max.nTuningRuns
## 1 Random.myrandom_classfr myrandom_classfr .rnorm 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.339 0.001 0.5007516
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.1 0.2867534 0.1673743
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.1565447 0.1786398 0 0.4909227
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.1 0.2865473 0.1672338
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.1513467 0.1840753 0# Any models that have tuning parameters has "better" results with cross-validation
# (except rf) & "different" results for different outcome metrics
# Max.cor.Y
# Check impact of cv
# rpart is not a good candidate since caret does not optimize cp (only tuning parameter of rpart) well
ret_lst <- myfit_mdl(model_id="Max.cor.Y.cv.0",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df)## [1] "fitting model: Max.cor.Y.cv.0.rpart"
## [1] " indep_vars: A.nuppr.log"## Loading required package: rpart## Fitting cp = 0 on full training set## Loading required package: rpart.plot## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 4475
##
## CP nsplit rel error
## 1 0 0 1
##
## Node number 1: 4475 observations
## predicted class=N expected loss=0.1673743 P(node) =1
## class counts: 3726 749
## probabilities: 0.833 0.167
##
## n= 4475
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 4475 749 N (0.8326257 0.1673743) *
## [1] " calling mypredict_mdl for fit:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 3726
## 2 Y 749
## Prediction
## Reference N Y
## N 3726 0
## Y 749 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.326257e-01 0.000000e+00 8.213602e-01 8.434553e-01 8.326257e-01
## AccuracyPValue McnemarPValue
## 5.097571e-01 1.800616e-164
## [1] " calling mypredict_mdl for OOB:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 1713
## 2 Y 344
## Prediction
## Reference N Y
## N 1713 0
## Y 344 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.327662e-01 0.000000e+00 8.159247e-01 8.486533e-01 8.327662e-01
## AccuracyPValue McnemarPValue
## 5.143944e-01 2.337097e-76
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.cv.0.rpart rpart A.nuppr.log 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.65 0.057 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.8326257
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8213602 0.8434553 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.8327662
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8159247 0.8486533 0ret_lst <- myfit_mdl(model_id="Max.cor.Y.cv.0.cp.0",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df,
n_cv_folds=0,
tune_models_df=data.frame(parameter="cp", min=0.0, max=0.0, by=0.1))## [1] "fitting model: Max.cor.Y.cv.0.cp.0.rpart"
## [1] " indep_vars: A.nuppr.log"
## Fitting cp = 0 on full training set
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 4475
##
## CP nsplit rel error
## 1 0 0 1
##
## Node number 1: 4475 observations
## predicted class=N expected loss=0.1673743 P(node) =1
## class counts: 3726 749
## probabilities: 0.833 0.167
##
## n= 4475
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 4475 749 N (0.8326257 0.1673743) *
## [1] " calling mypredict_mdl for fit:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 3726
## 2 Y 749
## Prediction
## Reference N Y
## N 3726 0
## Y 749 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.326257e-01 0.000000e+00 8.213602e-01 8.434553e-01 8.326257e-01
## AccuracyPValue McnemarPValue
## 5.097571e-01 1.800616e-164
## [1] " calling mypredict_mdl for OOB:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 1713
## 2 Y 344
## Prediction
## Reference N Y
## N 1713 0
## Y 344 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.327662e-01 0.000000e+00 8.159247e-01 8.486533e-01 8.327662e-01
## AccuracyPValue McnemarPValue
## 5.143944e-01 2.337097e-76
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.cv.0.cp.0.rpart rpart A.nuppr.log 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.596 0.056 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.8326257
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8213602 0.8434553 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.8327662
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8159247 0.8486533 0if (glb_is_regression || glb_is_binomial) # For multinomials this model will be run next by default
ret_lst <- myfit_mdl(model_id="Max.cor.Y",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Max.cor.Y.rpart"
## [1] " indep_vars: A.nuppr.log"
## Aggregating results
## Fitting final model on full training set## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 4475
##
## CP nsplit rel error
## 1 0 0 1
##
## Node number 1: 4475 observations
## predicted class=N expected loss=0.1673743 P(node) =1
## class counts: 3726 749
## probabilities: 0.833 0.167
##
## n= 4475
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 4475 749 N (0.8326257 0.1673743) *
## [1] " calling mypredict_mdl for fit:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.Max.cor.Y.rpart.N
## 1 N 3726
## 2 Y 749
## Prediction
## Reference N Y
## N 3726 0
## Y 749 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.326257e-01 0.000000e+00 8.213602e-01 8.434553e-01 8.326257e-01
## AccuracyPValue McnemarPValue
## 5.097571e-01 1.800616e-164
## [1] " calling mypredict_mdl for OOB:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.Max.cor.Y.rpart.N
## 1 N 1713
## 2 Y 344
## Prediction
## Reference N Y
## N 1713 0
## Y 344 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.327662e-01 0.000000e+00 8.159247e-01 8.486533e-01 8.327662e-01
## AccuracyPValue McnemarPValue
## 5.143944e-01 2.337097e-76
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.rpart rpart A.nuppr.log 1
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 1.274 0.056 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.8326258
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8213602 0.8434553 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.8327662
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8159247 0.8486533 0
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.0002791548 0# Used to compare vs. Interactions.High.cor.Y
ret_lst <- myfit_mdl(model_id="Max.cor.Y",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Max.cor.Y.glm"
## [1] " indep_vars: A.nuppr.log"
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.3585 -0.6318 -0.4867 -0.3464 2.6336
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.41620 0.11470 3.628 0.000285 ***
## A.nuppr.log -1.38947 0.08027 -17.310 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 4042.7 on 4474 degrees of freedom
## Residual deviance: 3710.6 on 4473 degrees of freedom
## AIC: 3714.6
##
## Number of Fisher Scoring iterations: 5
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.2867534
## 2 0.1 0.3499729
## 3 0.2 0.3986014
## 4 0.3 0.3121547
## 5 0.4 0.0000000
## 6 0.5 0.0000000
## 7 0.6 0.0000000
## 8 0.7 0.0000000
## 9 0.8 0.0000000
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.Max.cor.Y.glm.N
## 1 N 2872
## 2 Y 350
## Popular.fctr.predict.Max.cor.Y.glm.Y
## 1 854
## 2 399
## Prediction
## Reference N Y
## N 2872 854
## Y 350 399
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.309497e-01 2.392074e-01 7.176970e-01 7.439004e-01 8.326257e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.280095e-47
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.2865473
## 2 0.1 0.3485577
## 3 0.2 0.3880266
## 4 0.3 0.3465046
## 5 0.4 0.0000000
## 6 0.5 0.0000000
## 7 0.6 0.0000000
## 8 0.7 0.0000000
## 9 0.8 0.0000000
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.Max.cor.Y.glm.N
## 1 N 1330
## 2 Y 169
## Popular.fctr.predict.Max.cor.Y.glm.Y
## 1 383
## 2 175
## Prediction
## Reference N Y
## N 1330 383
## Y 169 175
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.316480e-01 2.283681e-01 7.119353e-01 7.506985e-01 8.327662e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.236001e-19
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.glm glm A.nuppr.log 1
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 1.212 0.079 0.7073742
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.2 0.3986014 0.8324022
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.717697 0.7439004 -0.0004459345 0.710206
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.3880266 0.731648
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.7119353 0.7506985 0.2283681 3714.601
## max.AccuracySD.fit max.KappaSD.fit
## 1 6.48833e-05 0.0007723812# Interactions.High.cor.Y
if (length(int_feats <- setdiff(unique(glb_feats_df$cor.high.X), NA)) > 0) {
# lm & glm handle interaction terms; rpart & rf do not
if (glb_is_regression || glb_is_binomial) {
indep_vars_vctr <-
c(max_cor_y_x_var, paste(max_cor_y_x_var, int_feats, sep=":"))
} else { indep_vars_vctr <- union(max_cor_y_x_var, int_feats) }
ret_lst <- myfit_mdl(model_id="Interact.High.cor.Y",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
model_type=glb_model_type,
indep_vars_vctr,
glb_rsp_var, glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)
} ## [1] "fitting model: Interact.High.cor.Y.glm"
## [1] " indep_vars: A.nuppr.log, A.nuppr.log:A.nwrds.log, A.nuppr.log:A.npnct21.log, A.nuppr.log:H.npnct09.log, A.nuppr.log:S.T.make, A.nuppr.log:H.npnct17.log, A.nuppr.log:S.T.can, A.nuppr.log:S.npnct01.log, A.nuppr.log:A.npnct23.log, A.nuppr.log:S.T.said, A.nuppr.log:A.npnct25.log, A.nuppr.log:S.T.one, A.nuppr.log:S.T.state, A.nuppr.log:S.npnct07.log, A.nuppr.log:A.npnct19.log, A.nuppr.log:S.npnct03.log, A.nuppr.log:A.npnct18.log, A.nuppr.log:A.npnct20.log, A.nuppr.log:S.has.year.colon, A.nuppr.log:S.npnct22.log, A.nuppr.log:S.T.take, A.nuppr.log:S.npnct06.log, A.nuppr.log:A.npnct02.log, A.nuppr.log:S.T.new, A.nuppr.log:S.npnct13.log, A.nuppr.log:A.T.will, A.nuppr.log:S.T.show, A.nuppr.log:S.T.day, A.nuppr.log:S.npnct30.log, A.nuppr.log:A.T.first, A.nuppr.log:A.T.year, A.nuppr.log:A.T.report, A.nuppr.log:A.T.compani, A.nuppr.log:A.npnct30.log, A.nuppr.log:S.T.share, A.nuppr.log:H.T.billion, A.nuppr.log:S.T.time, A.nuppr.log:A.npnct14.log, A.nuppr.log:A.T.articl, A.nuppr.log:A.T.newyork, A.nuppr.log:H.T.polit, A.nuppr.log:H.T.springsumm, A.nuppr.log:S.npnct04.log, A.nuppr.log:H.T.report, A.nuppr.log:H.npnct15.log, A.nuppr.log:S.npnct16.log, A.nuppr.log:A.T.intern, A.nuppr.log:S.T.intern, A.nuppr.log:H.T.X2015, A.nuppr.log:A.T.week, A.nuppr.log:S.npnct12.log, A.nuppr.log:S.ndgts.log, A.nuppr.log:H.nuppr.log, A.nuppr.log:A.nchrs.log, A.nuppr.log:S.nchrs.log, A.nuppr.log:S.nuppr.log"
## Aggregating results
## Fitting final model on full training set## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.70376 -0.66551 -0.36072 -0.09499 3.08743
##
## Coefficients: (2 not defined because of singularities)
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -5.316e-01 3.129e-01 -1.699 0.089398 .
## A.nuppr.log 4.948e-01 9.968e-01 0.496 0.619646
## `A.nuppr.log:A.nwrds.log` 9.983e-01 3.444e-01 2.899 0.003747 **
## `A.nuppr.log:A.npnct21.log` 6.032e-01 1.396e-01 4.320 1.56e-05 ***
## `A.nuppr.log:H.npnct09.log` 5.307e-01 3.591e-01 1.478 0.139431
## `A.nuppr.log:S.T.make` 3.258e-01 3.356e-01 0.971 0.331734
## `A.nuppr.log:H.npnct17.log` 7.791e-01 2.586e-01 3.013 0.002587 **
## `A.nuppr.log:S.T.can` -8.989e-02 3.941e-01 -0.228 0.819582
## `A.nuppr.log:S.npnct01.log` 9.274e-01 4.620e-01 2.007 0.044735 *
## `A.nuppr.log:A.npnct23.log` -8.718e+00 2.480e+03 -0.004 0.997195
## `A.nuppr.log:S.T.said` 1.195e+00 4.044e-01 2.955 0.003123 **
## `A.nuppr.log:A.npnct25.log` NA NA NA NA
## `A.nuppr.log:S.T.one` 9.074e-02 3.680e-01 0.247 0.805234
## `A.nuppr.log:S.T.state` 7.073e-01 3.184e-01 2.221 0.026330 *
## `A.nuppr.log:S.npnct07.log` -3.228e+01 3.503e+03 -0.009 0.992648
## `A.nuppr.log:A.npnct19.log` 9.579e+00 1.085e+05 0.000 0.999930
## `A.nuppr.log:S.npnct03.log` -1.114e+01 1.895e+03 -0.006 0.995310
## `A.nuppr.log:A.npnct18.log` -1.563e+01 6.784e+04 0.000 0.999816
## `A.nuppr.log:A.npnct20.log` NA NA NA NA
## `A.nuppr.log:S.has.year.colon` -1.038e+01 9.663e+02 -0.011 0.991429
## `A.nuppr.log:S.npnct22.log` -1.427e+01 1.733e+03 -0.008 0.993429
## `A.nuppr.log:S.T.take` -9.060e-01 5.779e-01 -1.568 0.116955
## `A.nuppr.log:S.npnct06.log` -7.721e-01 8.395e-01 -0.920 0.357737
## `A.nuppr.log:A.npnct02.log` -1.093e+01 9.051e+03 -0.001 0.999037
## `A.nuppr.log:S.T.new` -5.155e-01 4.103e-01 -1.256 0.208977
## `A.nuppr.log:S.npnct13.log` -9.163e-02 1.024e-01 -0.895 0.370873
## `A.nuppr.log:A.T.will` -1.038e+00 4.135e-01 -2.510 0.012058 *
## `A.nuppr.log:S.T.show` -1.591e+00 6.739e-01 -2.361 0.018205 *
## `A.nuppr.log:S.T.day` -7.395e-01 5.891e-01 -1.255 0.209311
## `A.nuppr.log:S.npnct30.log` -1.284e+01 6.239e+03 -0.002 0.998358
## `A.nuppr.log:A.T.first` -6.053e-01 5.042e-01 -1.200 0.229950
## `A.nuppr.log:A.T.year` -5.359e-01 5.280e-01 -1.015 0.310091
## `A.nuppr.log:A.T.report` -1.440e+00 7.148e-01 -2.014 0.043963 *
## `A.nuppr.log:A.T.compani` -1.402e+00 5.812e-01 -2.413 0.015835 *
## `A.nuppr.log:A.npnct30.log` 4.843e+00 6.136e+03 0.001 0.999370
## `A.nuppr.log:S.T.share` -1.376e+00 6.530e-01 -2.107 0.035156 *
## `A.nuppr.log:H.T.billion` -5.909e-01 5.647e-01 -1.046 0.295404
## `A.nuppr.log:S.T.time` -7.317e-02 3.913e-01 -0.187 0.851646
## `A.nuppr.log:A.npnct14.log` 7.553e-01 1.090e-01 6.929 4.23e-12 ***
## `A.nuppr.log:A.T.articl` -1.486e+00 6.941e-01 -2.140 0.032326 *
## `A.nuppr.log:A.T.newyork` 6.631e-01 3.550e-01 1.868 0.061782 .
## `A.nuppr.log:H.T.polit` -7.425e-01 3.002e-01 -2.474 0.013374 *
## `A.nuppr.log:H.T.springsumm` 2.066e+01 1.897e+03 0.011 0.991312
## `A.nuppr.log:S.npnct04.log` -1.098e+00 4.328e-01 -2.537 0.011177 *
## `A.nuppr.log:H.T.report` -1.170e+00 5.391e-01 -2.170 0.030014 *
## `A.nuppr.log:H.npnct15.log` -2.919e+01 2.063e+03 -0.014 0.988711
## `A.nuppr.log:S.npnct16.log` -2.692e-01 2.182e-01 -1.234 0.217346
## `A.nuppr.log:A.T.intern` 4.713e+02 7.523e+04 0.006 0.995002
## `A.nuppr.log:S.T.intern` -4.735e+02 7.523e+04 -0.006 0.994978
## `A.nuppr.log:H.T.X2015` -2.136e+01 1.211e+03 -0.018 0.985930
## `A.nuppr.log:A.T.week` -1.964e+00 5.375e-01 -3.653 0.000259 ***
## `A.nuppr.log:S.npnct12.log` -5.688e-02 6.728e-02 -0.845 0.397852
## `A.nuppr.log:S.ndgts.log` -2.435e-01 7.447e-02 -3.270 0.001075 **
## `A.nuppr.log:H.nuppr.log` -4.840e-01 9.772e-02 -4.953 7.30e-07 ***
## `A.nuppr.log:A.nchrs.log` 3.006e-03 2.726e+00 0.001 0.999120
## `A.nuppr.log:S.nchrs.log` -4.080e-01 2.733e+00 -0.149 0.881317
## `A.nuppr.log:S.nuppr.log` -5.704e-01 1.834e-01 -3.110 0.001872 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 4042.7 on 4474 degrees of freedom
## Residual deviance: 3289.6 on 4420 degrees of freedom
## AIC: 3399.6
##
## Number of Fisher Scoring iterations: 17
##
## [1] " calling mypredict_mdl for fit:"## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## threshold f.score
## 1 0.0 0.28675345
## 2 0.1 0.39963168
## 3 0.2 0.46502058
## 4 0.3 0.47787611
## 5 0.4 0.36851521
## 6 0.5 0.21552723
## 7 0.6 0.04150454
## 8 0.7 0.00794702
## 9 0.8 0.00000000
## 10 0.9 0.00000000
## 11 1.0 0.00000000## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 3185
## 2 Y 344
## Popular.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 541
## 2 405
## Prediction
## Reference N Y
## N 3185 541
## Y 344 405
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.022346e-01 3.579176e-01 7.902571e-01 8.138164e-01 8.326257e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 4.443882e-11
## [1] " calling mypredict_mdl for OOB:"## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## threshold f.score
## 1 0.0 0.286547272
## 2 0.1 0.395442359
## 3 0.2 0.448942042
## 4 0.3 0.459563543
## 5 0.4 0.375494071
## 6 0.5 0.187341772
## 7 0.6 0.067039106
## 8 0.7 0.017291066
## 9 0.8 0.005797101
## 10 0.9 0.000000000
## 11 1.0 0.000000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 1457
## 2 Y 165
## Popular.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 256
## 2 179
## Prediction
## Reference N Y
## N 1457 256
## Y 165 179
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.953330e-01 3.354449e-01 7.772394e-01 8.125808e-01 8.327662e-01
## AccuracyPValue McnemarPValue
## 9.999959e-01 1.152783e-05
## model_id model_method
## 1 Interact.High.cor.Y.glm glm
## feats
## 1 A.nuppr.log, A.nuppr.log:A.nwrds.log, A.nuppr.log:A.npnct21.log, A.nuppr.log:H.npnct09.log, A.nuppr.log:S.T.make, A.nuppr.log:H.npnct17.log, A.nuppr.log:S.T.can, A.nuppr.log:S.npnct01.log, A.nuppr.log:A.npnct23.log, A.nuppr.log:S.T.said, A.nuppr.log:A.npnct25.log, A.nuppr.log:S.T.one, A.nuppr.log:S.T.state, A.nuppr.log:S.npnct07.log, A.nuppr.log:A.npnct19.log, A.nuppr.log:S.npnct03.log, A.nuppr.log:A.npnct18.log, A.nuppr.log:A.npnct20.log, A.nuppr.log:S.has.year.colon, A.nuppr.log:S.npnct22.log, A.nuppr.log:S.T.take, A.nuppr.log:S.npnct06.log, A.nuppr.log:A.npnct02.log, A.nuppr.log:S.T.new, A.nuppr.log:S.npnct13.log, A.nuppr.log:A.T.will, A.nuppr.log:S.T.show, A.nuppr.log:S.T.day, A.nuppr.log:S.npnct30.log, A.nuppr.log:A.T.first, A.nuppr.log:A.T.year, A.nuppr.log:A.T.report, A.nuppr.log:A.T.compani, A.nuppr.log:A.npnct30.log, A.nuppr.log:S.T.share, A.nuppr.log:H.T.billion, A.nuppr.log:S.T.time, A.nuppr.log:A.npnct14.log, A.nuppr.log:A.T.articl, A.nuppr.log:A.T.newyork, A.nuppr.log:H.T.polit, A.nuppr.log:H.T.springsumm, A.nuppr.log:S.npnct04.log, A.nuppr.log:H.T.report, A.nuppr.log:H.npnct15.log, A.nuppr.log:S.npnct16.log, A.nuppr.log:A.T.intern, A.nuppr.log:S.T.intern, A.nuppr.log:H.T.X2015, A.nuppr.log:A.T.week, A.nuppr.log:S.npnct12.log, A.nuppr.log:S.ndgts.log, A.nuppr.log:H.nuppr.log, A.nuppr.log:A.nchrs.log, A.nuppr.log:S.nchrs.log, A.nuppr.log:S.nuppr.log
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 3.104 1.108
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.7990119 0.3 0.4778761 0.8451393
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7902571 0.8138164 0.1744465 0.7758607
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.4595635 0.795333
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.7772394 0.8125808 0.3354449 3399.63
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.004202829 0.01704351# Low.cor.X
# if (glb_is_classification && glb_is_binomial)
# indep_vars_vctr <- subset(glb_feats_df, is.na(cor.high.X) &
# is.ConditionalX.y &
# (exclude.as.feat != 1))[, "id"] else
indep_vars_vctr <- subset(glb_feats_df, is.na(cor.high.X) & !myNearZV &
(exclude.as.feat != 1))[, "id"]
ret_lst <- myfit_mdl(model_id="Low.cor.X",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
glb_rsp_var, glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Low.cor.X.glm"
## [1] " indep_vars: WordCount.log, H.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, A.npnct21.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, S.T.make, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, S.T.one, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, .rnorm, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, H.T.big, A.npnct20.log, S.has.year.colon, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, S.npnct06.log, H.npnct14.log, H.T.deal, S.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, H.T.new, S.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, A.T.year, A.T.report, A.T.compani, H.T.busi, A.npnct14.log, S.T.share, S.T.time, A.T.articl, H.T.newyork, A.T.newyork, H.T.springsumm, H.T.day, H.T.report, S.npnct04.log, S.npnct16.log, H.T.week, A.T.intern, H.T.fashion, H.npnct16.log, A.T.fashion, A.T.week, H.npnct30.log, S.npnct12.log, H.ndgts.log, S.ndgts.log, H.nuppr.log, H.nchrs.log, A.nchrs.log, A.nwrds.unq.log, S.nuppr.log"
## Aggregating results
## Fitting final model on full training set## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning: not plotting observations with leverage one:
## 2501
## Warning: not plotting observations with leverage one:
## 2501
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.8204 -0.3159 -0.1312 0.0000 3.6505
##
## Coefficients: (3 not defined because of singularities)
## Estimate
## (Intercept) -2.748e+00
## WordCount.log 1.127e+00
## H.nwrds.log 5.287e-01
## `PubDate.hour.fctr(7.67,15.3]` 2.343e-01
## `PubDate.hour.fctr(15.3,23]` 4.066e-01
## A.nwrds.log -9.763e-01
## H.npnct21.log 1.488e+00
## PubDate.wkend -3.372e-01
## A.npnct21.log 1.435e+00
## H.npnct09.log 1.945e+00
## PubDate.last10.log 2.252e-01
## PubDate.last1.log -4.492e-02
## S.T.make -1.135e+00
## S.npnct01.log 2.001e+00
## S.T.can -1.274e+00
## H.npnct17.log 1.060e+00
## H.has.ebola -2.699e-01
## H.npnct01.log -1.496e+00
## S.T.said 1.473e+00
## H.T.make -1.680e-01
## H.npnct12.log 4.681e-01
## `myCategory.fctrForeign#World#Asia Pacific` -3.838e+00
## `myCategory.fctr#Multimedia#` -4.232e+00
## `myCategory.fctrCulture#Arts#` -2.823e+00
## `myCategory.fctrBusiness#Business Day#Dealbook` -2.536e+00
## myCategory.fctrmyOther -2.117e+01
## `myCategory.fctrBusiness#Technology#` -1.867e+00
## `myCategory.fctrBusiness#Crosswords/Games#` 8.317e-01
## `myCategory.fctrTStyle##` -4.344e+00
## `myCategory.fctrForeign#World#` -1.883e+01
## `myCategory.fctrOpEd#Opinion#` 6.538e-01
## `myCategory.fctrStyles##Fashion` -2.269e+01
## `myCategory.fctr#Opinion#Room For Debate` -5.785e+00
## `myCategory.fctr#U.S.#Education` -2.168e+01
## `myCategory.fctr##` -2.624e+00
## `myCategory.fctrMetro#N.Y. / Region#` -2.037e+00
## `myCategory.fctrBusiness#Business Day#Small Business` -4.395e+00
## `myCategory.fctrStyles#U.S.#` -4.788e-01
## `myCategory.fctrTravel#Travel#` -4.023e+00
## `myCategory.fctr#Opinion#The Public Editor` 1.075e+00
## S.T.one -1.037e+00
## S.T.state 1.478e+00
## H.T.take -2.639e-01
## A.npnct17.log -1.698e-01
## S.npnct17.log NA
## A.T.presid 5.367e+02
## S.T.presid -5.362e+02
## S.npnct08.log 1.379e+01
## H.T.time 1.491e-01
## A.npnct08.log NA
## S.npnct09.log -1.233e+01
## A.npnct09.log NA
## PubDate.last100.log 1.446e-02
## .rnorm -7.241e-02
## H.npnct05.log -2.532e+01
## H.T.obama -1.444e-01
## H.T.say -6.471e-01
## H.T.bank 2.742e-01
## `PubDate.date.fctr(7,13]` -6.259e-02
## `PubDate.date.fctr(13,19]` -1.762e-01
## `PubDate.date.fctr(19,25]` -1.147e-01
## `PubDate.date.fctr(25,31]` 1.184e-01
## `PubDate.second.fctr(14.8,29.5]` 1.115e-01
## `PubDate.second.fctr(29.5,44.2]` 1.025e-03
## `PubDate.second.fctr(44.2,59.1]` -2.817e-01
## H.npnct07.log 2.272e-01
## S.npnct07.log -3.359e+01
## S.npnct03.log -2.905e+01
## A.npnct19.log -2.303e+01
## H.npnct13.log 3.531e-01
## H.T.big -4.463e-01
## A.npnct20.log -3.310e+00
## S.has.year.colon -1.323e+01
## S.npnct22.log -2.488e+01
## H.npnct02.log -1.826e+01
## H.T.test -1.230e-01
## S.npnct15.log 3.617e-01
## S.T.take -5.749e-01
## S.npnct06.log 1.534e-01
## H.npnct14.log -1.969e-01
## H.T.deal -2.294e+01
## S.T.new 5.514e-02
## H.T.billion -3.929e-01
## H.T.polit -8.283e-01
## H.T.china -7.701e-01
## H.T.art -1.001e+00
## `PubDate.minute.fctr(14.8,29.5]` -1.444e-01
## `PubDate.minute.fctr(29.5,44.2]` -2.290e-01
## `PubDate.minute.fctr(44.2,59.1]` 1.875e-03
## S.npnct13.log -1.692e-01
## A.T.will -1.093e+00
## PubDate.wkday.fctr1 -6.483e-01
## PubDate.wkday.fctr2 -1.251e+00
## PubDate.wkday.fctr3 -8.655e-01
## PubDate.wkday.fctr4 -1.092e+00
## PubDate.wkday.fctr5 -9.206e-01
## PubDate.wkday.fctr6 -1.272e+00
## H.T.pictur 2.972e-01
## S.T.day -5.547e-01
## S.T.show -1.515e+00
## H.T.new -4.940e-01
## S.npnct30.log -1.513e+01
## H.T.news -5.654e-01
## H.T.first -9.638e-01
## H.T.X2014 -3.545e-01
## A.T.first -2.167e-01
## A.T.year -7.364e-01
## A.T.report -2.921e+00
## A.T.compani -9.640e-01
## H.T.busi -6.922e-01
## A.npnct14.log 8.924e-01
## S.T.share -1.700e+00
## S.T.time -1.078e+00
## A.T.articl -6.541e-01
## H.T.newyork -7.851e-01
## A.T.newyork 2.404e+00
## H.T.springsumm -1.732e+01
## H.T.day -4.617e-01
## H.T.report -1.326e+00
## S.npnct04.log -1.164e+00
## S.npnct16.log 1.580e-01
## H.T.week -3.672e-01
## A.T.intern -2.783e+00
## H.T.fashion 2.117e+00
## H.npnct16.log -4.049e-01
## A.T.fashion -5.294e+01
## A.T.week -1.209e+00
## H.npnct30.log -1.121e-01
## S.npnct12.log -1.786e-01
## H.ndgts.log 2.950e-01
## S.ndgts.log -3.106e-01
## H.nuppr.log 1.115e+00
## H.nchrs.log -1.433e+00
## A.nchrs.log 5.040e-01
## A.nwrds.unq.log -1.111e+00
## S.nuppr.log -6.250e-01
## Std. Error z value
## (Intercept) 2.988e+00 -0.920
## WordCount.log 9.057e-02 12.444
## H.nwrds.log 6.890e-01 0.767
## `PubDate.hour.fctr(7.67,15.3]` 2.454e-01 0.955
## `PubDate.hour.fctr(15.3,23]` 2.486e-01 1.636
## A.nwrds.log 8.029e-01 -1.216
## H.npnct21.log 3.166e-01 4.701
## PubDate.wkend 4.501e-01 -0.749
## A.npnct21.log 3.318e-01 4.325
## H.npnct09.log 7.141e-01 2.724
## PubDate.last10.log 1.245e-01 1.809
## PubDate.last1.log 4.431e-02 -1.014
## S.T.make 6.030e-01 -1.882
## S.npnct01.log 1.838e+00 1.089
## S.T.can 8.334e-01 -1.528
## H.npnct17.log 5.751e-01 1.843
## H.has.ebola 4.468e-01 -0.604
## H.npnct01.log 1.264e+00 -1.183
## S.T.said 7.783e-01 1.892
## H.T.make 3.354e-01 -0.501
## H.npnct12.log 2.093e-01 2.236
## `myCategory.fctrForeign#World#Asia Pacific` 6.885e-01 -5.575
## `myCategory.fctr#Multimedia#` 8.108e-01 -5.220
## `myCategory.fctrCulture#Arts#` 3.668e-01 -7.697
## `myCategory.fctrBusiness#Business Day#Dealbook` 3.157e-01 -8.034
## myCategory.fctrmyOther 3.027e+03 -0.007
## `myCategory.fctrBusiness#Technology#` 3.268e-01 -5.713
## `myCategory.fctrBusiness#Crosswords/Games#` 5.119e-01 1.625
## `myCategory.fctrTStyle##` 5.226e-01 -8.312
## `myCategory.fctrForeign#World#` 1.402e+03 -0.013
## `myCategory.fctrOpEd#Opinion#` 2.980e-01 2.194
## `myCategory.fctrStyles##Fashion` 1.487e+03 -0.015
## `myCategory.fctr#Opinion#Room For Debate` 6.357e-01 -9.099
## `myCategory.fctr#U.S.#Education` 9.806e+02 -0.022
## `myCategory.fctr##` 2.935e-01 -8.942
## `myCategory.fctrMetro#N.Y. / Region#` 4.855e-01 -4.196
## `myCategory.fctrBusiness#Business Day#Small Business` 7.101e-01 -6.189
## `myCategory.fctrStyles#U.S.#` 3.358e-01 -1.426
## `myCategory.fctrTravel#Travel#` 1.048e+00 -3.837
## `myCategory.fctr#Opinion#The Public Editor` 1.224e+00 0.878
## S.T.one 6.057e-01 -1.712
## S.T.state 8.398e-01 1.760
## H.T.take 4.617e-01 -0.572
## A.npnct17.log 1.347e+00 -0.126
## S.npnct17.log NA NA
## A.T.presid 1.572e+05 0.003
## S.T.presid 1.572e+05 -0.003
## S.npnct08.log 1.279e+04 0.001
## H.T.time 3.019e-01 0.494
## A.npnct08.log NA NA
## S.npnct09.log 1.279e+04 -0.001
## A.npnct09.log NA NA
## PubDate.last100.log 4.532e-02 0.319
## .rnorm 6.298e-02 -1.150
## H.npnct05.log 1.030e+04 -0.002
## H.T.obama 4.409e-01 -0.328
## H.T.say 4.253e-01 -1.522
## H.T.bank 4.816e-01 0.569
## `PubDate.date.fctr(7,13]` 1.968e-01 -0.318
## `PubDate.date.fctr(13,19]` 1.939e-01 -0.909
## `PubDate.date.fctr(19,25]` 1.917e-01 -0.598
## `PubDate.date.fctr(25,31]` 2.062e-01 0.574
## `PubDate.second.fctr(14.8,29.5]` 1.748e-01 0.638
## `PubDate.second.fctr(29.5,44.2]` 1.711e-01 0.006
## `PubDate.second.fctr(44.2,59.1]` 1.778e-01 -1.584
## H.npnct07.log 1.853e-01 1.226
## S.npnct07.log 9.747e+03 -0.003
## S.npnct03.log 8.794e+03 -0.003
## A.npnct19.log 4.681e+04 0.000
## H.npnct13.log 3.115e-01 1.134
## H.T.big 5.970e-01 -0.748
## A.npnct20.log 2.891e+04 0.000
## S.has.year.colon 4.978e+03 -0.003
## S.npnct22.log 7.190e+03 -0.003
## H.npnct02.log 5.027e+03 -0.004
## H.T.test 7.093e-01 -0.173
## S.npnct15.log 1.488e+00 0.243
## S.T.take 1.016e+00 -0.566
## S.npnct06.log 1.549e+00 0.099
## H.npnct14.log 1.970e-01 -1.000
## H.T.deal 2.771e+03 -0.008
## S.T.new 7.210e-01 0.076
## H.T.billion 8.425e-01 -0.466
## H.T.polit 3.243e-01 -2.554
## H.T.china 9.744e-01 -0.790
## H.T.art 8.356e-01 -1.198
## `PubDate.minute.fctr(14.8,29.5]` 1.815e-01 -0.796
## `PubDate.minute.fctr(29.5,44.2]` 1.766e-01 -1.296
## `PubDate.minute.fctr(44.2,59.1]` 1.813e-01 0.010
## S.npnct13.log 2.026e-01 -0.835
## A.T.will 7.776e-01 -1.405
## PubDate.wkday.fctr1 5.264e-01 -1.232
## PubDate.wkday.fctr2 5.741e-01 -2.179
## PubDate.wkday.fctr3 5.659e-01 -1.529
## PubDate.wkday.fctr4 5.599e-01 -1.951
## PubDate.wkday.fctr5 5.662e-01 -1.626
## PubDate.wkday.fctr6 4.718e-01 -2.695
## H.T.pictur 6.365e-01 0.467
## S.T.day 9.902e-01 -0.560
## S.T.show 1.117e+00 -1.356
## H.T.new 4.799e-01 -1.029
## S.npnct30.log 2.101e+03 -0.007
## H.T.news 8.223e-01 -0.688
## H.T.first 9.930e-01 -0.971
## H.T.X2014 9.690e-01 -0.366
## A.T.first 1.032e+00 -0.210
## A.T.year 9.576e-01 -0.769
## A.T.report 1.211e+00 -2.412
## A.T.compani 9.255e-01 -1.042
## H.T.busi 7.758e-01 -0.892
## A.npnct14.log 2.594e-01 3.440
## S.T.share 1.039e+00 -1.636
## S.T.time 9.376e-01 -1.150
## A.T.articl 1.915e+00 -0.342
## H.T.newyork 4.970e-01 -1.580
## A.T.newyork 9.848e-01 2.441
## H.T.springsumm 1.589e+03 -0.011
## H.T.day 5.816e-01 -0.794
## H.T.report 8.203e-01 -1.617
## S.npnct04.log 6.808e-01 -1.710
## S.npnct16.log 4.908e-01 0.322
## H.T.week 5.517e-01 -0.666
## A.T.intern 2.640e+00 -1.054
## H.T.fashion 1.478e+00 1.432
## H.npnct16.log 2.864e-01 -1.414
## A.T.fashion 2.505e+03 -0.021
## A.T.week 8.310e-01 -1.454
## H.npnct30.log 1.698e+00 -0.066
## S.npnct12.log 1.445e-01 -1.236
## H.ndgts.log 2.290e-01 1.288
## S.ndgts.log 1.539e-01 -2.018
## H.nuppr.log 4.205e-01 2.652
## H.nchrs.log 3.596e-01 -3.984
## A.nchrs.log 4.909e-01 1.027
## A.nwrds.unq.log 5.515e-01 -2.015
## S.nuppr.log 1.551e-01 -4.030
## Pr(>|z|)
## (Intercept) 0.357701
## WordCount.log < 2e-16 ***
## H.nwrds.log 0.442895
## `PubDate.hour.fctr(7.67,15.3]` 0.339763
## `PubDate.hour.fctr(15.3,23]` 0.101890
## A.nwrds.log 0.224007
## H.npnct21.log 2.59e-06 ***
## PubDate.wkend 0.453664
## A.npnct21.log 1.53e-05 ***
## H.npnct09.log 0.006447 **
## PubDate.last10.log 0.070513 .
## PubDate.last1.log 0.310622
## S.T.make 0.059858 .
## S.npnct01.log 0.276171
## S.T.can 0.126439
## H.npnct17.log 0.065363 .
## H.has.ebola 0.545777
## H.npnct01.log 0.236626
## S.T.said 0.058431 .
## H.T.make 0.616373
## H.npnct12.log 0.025343 *
## `myCategory.fctrForeign#World#Asia Pacific` 2.47e-08 ***
## `myCategory.fctr#Multimedia#` 1.79e-07 ***
## `myCategory.fctrCulture#Arts#` 1.39e-14 ***
## `myCategory.fctrBusiness#Business Day#Dealbook` 9.44e-16 ***
## myCategory.fctrmyOther 0.994419
## `myCategory.fctrBusiness#Technology#` 1.11e-08 ***
## `myCategory.fctrBusiness#Crosswords/Games#` 0.104210
## `myCategory.fctrTStyle##` < 2e-16 ***
## `myCategory.fctrForeign#World#` 0.989285
## `myCategory.fctrOpEd#Opinion#` 0.028231 *
## `myCategory.fctrStyles##Fashion` 0.987826
## `myCategory.fctr#Opinion#Room For Debate` < 2e-16 ***
## `myCategory.fctr#U.S.#Education` 0.982357
## `myCategory.fctr##` < 2e-16 ***
## `myCategory.fctrMetro#N.Y. / Region#` 2.72e-05 ***
## `myCategory.fctrBusiness#Business Day#Small Business` 6.04e-10 ***
## `myCategory.fctrStyles#U.S.#` 0.153955
## `myCategory.fctrTravel#Travel#` 0.000124 ***
## `myCategory.fctr#Opinion#The Public Editor` 0.380018
## S.T.one 0.086839 .
## S.T.state 0.078395 .
## H.T.take 0.567613
## A.npnct17.log 0.899679
## S.npnct17.log NA
## A.T.presid 0.997275
## S.T.presid 0.997278
## S.npnct08.log 0.999140
## H.T.time 0.621318
## A.npnct08.log NA
## S.npnct09.log 0.999231
## A.npnct09.log NA
## PubDate.last100.log 0.749746
## .rnorm 0.250250
## H.npnct05.log 0.998038
## H.T.obama 0.743216
## H.T.say 0.128070
## H.T.bank 0.569143
## `PubDate.date.fctr(7,13]` 0.750494
## `PubDate.date.fctr(13,19]` 0.363555
## `PubDate.date.fctr(19,25]` 0.549773
## `PubDate.date.fctr(25,31]` 0.565707
## `PubDate.second.fctr(14.8,29.5]` 0.523613
## `PubDate.second.fctr(29.5,44.2]` 0.995222
## `PubDate.second.fctr(44.2,59.1]` 0.113090
## H.npnct07.log 0.220030
## S.npnct07.log 0.997250
## S.npnct03.log 0.997364
## A.npnct19.log 0.999608
## H.npnct13.log 0.256931
## H.T.big 0.454674
## A.npnct20.log 0.999909
## S.has.year.colon 0.997879
## S.npnct22.log 0.997239
## H.npnct02.log 0.997102
## H.T.test 0.862283
## S.npnct15.log 0.807992
## S.T.take 0.571550
## S.npnct06.log 0.921110
## H.npnct14.log 0.317493
## H.T.deal 0.993395
## S.T.new 0.939036
## H.T.billion 0.640945
## H.T.polit 0.010639 *
## H.T.china 0.429307
## H.T.art 0.231053
## `PubDate.minute.fctr(14.8,29.5]` 0.426227
## `PubDate.minute.fctr(29.5,44.2]` 0.194882
## `PubDate.minute.fctr(44.2,59.1]` 0.991751
## S.npnct13.log 0.403614
## A.T.will 0.159909
## PubDate.wkday.fctr1 0.218114
## PubDate.wkday.fctr2 0.029324 *
## PubDate.wkday.fctr3 0.126171
## PubDate.wkday.fctr4 0.051047 .
## PubDate.wkday.fctr5 0.103932
## PubDate.wkday.fctr6 0.007038 **
## H.T.pictur 0.640490
## S.T.day 0.575321
## S.T.show 0.174976
## H.T.new 0.303333
## S.npnct30.log 0.994257
## H.T.news 0.491721
## H.T.first 0.331766
## H.T.X2014 0.714511
## A.T.first 0.833697
## A.T.year 0.441870
## A.T.report 0.015855 *
## A.T.compani 0.297572
## H.T.busi 0.372281
## A.npnct14.log 0.000581 ***
## S.T.share 0.101745
## S.T.time 0.250299
## A.T.articl 0.732637
## H.T.newyork 0.114193
## A.T.newyork 0.014657 *
## H.T.springsumm 0.991302
## H.T.day 0.427300
## H.T.report 0.105934
## S.npnct04.log 0.087299 .
## S.npnct16.log 0.747427
## H.T.week 0.505725
## A.T.intern 0.291668
## H.T.fashion 0.152005
## H.npnct16.log 0.157409
## A.T.fashion 0.983137
## A.T.week 0.145880
## H.npnct30.log 0.947362
## S.npnct12.log 0.216499
## H.ndgts.log 0.197683
## S.ndgts.log 0.043615 *
## H.nuppr.log 0.007997 **
## H.nchrs.log 6.78e-05 ***
## A.nchrs.log 0.304604
## A.nwrds.unq.log 0.043897 *
## S.nuppr.log 5.57e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 4042.7 on 4474 degrees of freedom
## Residual deviance: 1833.4 on 4342 degrees of freedom
## AIC: 2099.4
##
## Number of Fisher Scoring iterations: 19
##
## [1] " calling mypredict_mdl for fit:"## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## threshold f.score
## 1 0.0 0.2867534
## 2 0.1 0.6610090
## 3 0.2 0.7301408
## 4 0.3 0.7445072
## 5 0.4 0.7405914
## 6 0.5 0.7307968
## 7 0.6 0.7037319
## 8 0.7 0.6611842
## 9 0.8 0.5683060
## 10 0.9 0.3995816
## 11 1.0 0.0000000## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.Low.cor.X.glm.N
## 1 N 3475
## 2 Y 156
## Popular.fctr.predict.Low.cor.X.glm.Y
## 1 251
## 2 593
## Prediction
## Reference N Y
## N 3475 251
## Y 156 593
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 9.090503e-01 6.894249e-01 9.002448e-01 9.173176e-01 8.326257e-01
## AccuracyPValue McnemarPValue
## 1.896445e-49 3.171216e-06
## [1] " calling mypredict_mdl for OOB:"## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## threshold f.score
## 1 0.0 0.2865473
## 2 0.1 0.6328852
## 3 0.2 0.7020202
## 4 0.3 0.7340720
## 5 0.4 0.7064083
## 6 0.5 0.6984127
## 7 0.6 0.6552316
## 8 0.7 0.6123188
## 9 0.8 0.5461690
## 10 0.9 0.3222749
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.Low.cor.X.glm.N
## 1 N 1600
## 2 Y 79
## Popular.fctr.predict.Low.cor.X.glm.Y
## 1 113
## 2 265
## Prediction
## Reference N Y
## N 1600 113
## Y 79 265
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 9.066602e-01 6.776204e-01 8.932593e-01 9.188884e-01 8.327662e-01
## AccuracyPValue McnemarPValue
## 3.600992e-22 1.723902e-02
## model_id model_method
## 1 Low.cor.X.glm glm
## feats
## 1 WordCount.log, H.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, A.npnct21.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, S.T.make, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, S.T.one, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, .rnorm, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, H.T.big, A.npnct20.log, S.has.year.colon, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, S.npnct06.log, H.npnct14.log, H.T.deal, S.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, H.T.new, S.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, A.T.year, A.T.report, A.T.compani, H.T.busi, A.npnct14.log, S.T.share, S.T.time, A.T.articl, H.T.newyork, A.T.newyork, H.T.springsumm, H.T.day, H.T.report, S.npnct04.log, S.npnct16.log, H.T.week, A.T.intern, H.T.fashion, H.npnct16.log, A.T.fashion, A.T.week, H.npnct30.log, S.npnct12.log, H.ndgts.log, S.ndgts.log, H.nuppr.log, H.nchrs.log, A.nchrs.log, A.nwrds.unq.log, S.nuppr.log
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 7.723 3.727
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.9494083 0.3 0.7445072 0.9061454
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.9002448 0.9173176 0.6418071 0.9160473
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.734072 0.9066602
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.8932593 0.9188884 0.6776204 2099.443
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.001143281 0.008381247glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 10 fit.models 7 0 236.209 275.696 39.487
## 11 fit.models 7 1 275.697 NA NAfit.models_1_chunk_df <- myadd_chunk(NULL, "fit.models_1_bgn")## label step_major step_minor bgn end elapsed
## 1 fit.models_1_bgn 1 0 279.819 NA NA# Options:
# 1. rpart & rf manual tuning
# 2. rf without pca (default: with pca)
# All X that is not user excluded
# if (glb_is_classification && glb_is_binomial) {
# model_id_pfx <- "Conditional.X"
# # indep_vars_vctr <- setdiff(names(glb_fitent_df), union(glb_rsp_var, glb_exclude_vars_as_features))
# indep_vars_vctr <- subset(glb_feats_df, is.ConditionalX.y &
# (exclude.as.feat != 1))[, "id"]
# } else {
model_id_pfx <- "All.X"
indep_vars_vctr <- subset(glb_feats_df, !myNearZV &
(exclude.as.feat != 1))[, "id"]
# }
for (method in glb_models_method_vctr) {
fit.models_1_chunk_df <- myadd_chunk(fit.models_1_chunk_df,
paste0("fit.models_1_", method), major.inc=TRUE)
if (method %in% c("rpart", "rf")) {
# rpart: fubar's the tree
# rf: skip the scenario w/ .rnorm for speed
indep_vars_vctr <- setdiff(indep_vars_vctr, c(".rnorm"))
model_id <- paste0(model_id_pfx, ".no.rnorm")
} else model_id <- model_id_pfx
ret_lst <- myfit_mdl(model_id=model_id, model_method=method,
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitent_df, OOB_df=glb_OOBent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df)
# Since caret does not optimize rpart well
# if (method == "rpart")
# ret_lst <- myfit_mdl(model_id=paste0(model_id_pfx, ".cp.0"), model_method=method,
# indep_vars_vctr=indep_vars_vctr,
# model_type=glb_model_type,
# rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
# fit_df=glb_fitent_df, OOB_df=glb_OOBent_df,
# n_cv_folds=0, tune_models_df=data.frame(parameter="cp", min=0.0, max=0.0, by=0.1))
}## label step_major step_minor bgn end elapsed
## 1 fit.models_1_bgn 1 0 279.819 279.831 0.012
## 2 fit.models_1_glm 2 0 279.831 NA NA
## [1] "fitting model: All.X.glm"
## [1] " indep_vars: WordCount.log, H.nwrds.log, S.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, S.npnct21.log, A.npnct21.log, H.npnct08.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, A.T.make, S.T.make, H.npnct06.log, A.T.can, A.npnct01.log, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, A.T.said, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, A.T.one, S.T.one, A.T.state, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, .rnorm, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, A.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, A.has.http, A.npnct03.log, H.T.big, A.npnct02.log, A.npnct18.log, A.npnct20.log, A.has.year.colon, S.has.year.colon, A.npnct22.log, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, A.T.take, A.npnct06.log, S.npnct06.log, A.npnct15.log, H.npnct14.log, H.T.deal, S.T.new, A.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.npnct13.log, A.T.will, S.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, A.T.show, A.T.day, H.T.new, S.npnct30.log, A.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, S.T.first, A.T.year, S.T.year, A.T.report, S.T.report, A.T.compani, S.T.compani, H.T.morn, H.T.busi, A.npnct14.log, A.T.share, S.T.share, H.npnct04.log, S.T.time, A.T.time, S.npnct14.log, A.T.articl, S.T.articl, H.T.newyork, A.T.newyork, S.T.newyork, H.T.today, H.T.springsumm, H.T.day, H.npnct15.log, H.T.report, A.npnct04.log, S.npnct04.log, H.T.daili, H.T.X2015, S.npnct16.log, H.T.week, A.npnct16.log, A.T.intern, S.T.intern, H.has.year.colon, H.T.fashion, H.npnct16.log, A.T.fashion, S.T.fashion, A.T.week, S.T.week, H.npnct30.log, S.npnct12.log, A.npnct12.log, H.ndgts.log, S.ndgts.log, A.ndgts.log, H.nuppr.log, H.nchrs.log, H.nwrds.unq.log, A.nchrs.log, S.nchrs.log, A.nwrds.unq.log, S.nwrds.unq.log, S.nuppr.log, A.nuppr.log"
## Aggregating results
## Fitting final model on full training set## Warning: glm.fit: algorithm did not converge## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning: not plotting observations with leverage one:
## 129, 1143, 2144, 2750, 3288, 3299, 3400, 3637, 3918, 3953, 4105
## Warning: not plotting observations with leverage one:
## 129, 1143, 2144, 2750, 3288, 3299, 3400, 3637, 3918, 3953, 4105
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -8.49 0.00 0.00 0.00 8.49
##
## Coefficients: (14 not defined because of singularities)
## Estimate
## (Intercept) -7.540e+14
## WordCount.log 3.185e+14
## H.nwrds.log -4.013e+13
## S.nwrds.log -2.810e+15
## `PubDate.hour.fctr(7.67,15.3]` -1.736e+13
## `PubDate.hour.fctr(15.3,23]` -6.108e+13
## A.nwrds.log 2.605e+15
## H.npnct21.log 5.271e+14
## PubDate.wkend -2.661e+14
## S.npnct21.log 1.043e+15
## A.npnct21.log -8.701e+14
## H.npnct08.log 1.233e+14
## H.npnct09.log NA
## PubDate.last10.log 7.101e+13
## PubDate.last1.log -1.649e+13
## A.T.make -5.708e+16
## S.T.make 5.676e+16
## H.npnct06.log 8.517e+14
## A.T.can -1.452e+15
## A.npnct01.log -8.624e+13
## S.npnct01.log NA
## S.T.can 1.530e+15
## H.npnct17.log -6.449e+14
## H.has.ebola 1.178e+14
## H.npnct01.log -8.440e+14
## A.T.said 8.275e+16
## S.T.said -8.253e+16
## H.T.make 1.903e+13
## H.npnct12.log 1.704e+14
## `myCategory.fctrForeign#World#Asia Pacific` -3.018e+15
## `myCategory.fctr#Multimedia#` -4.277e+14
## `myCategory.fctrCulture#Arts#` -2.828e+14
## `myCategory.fctrBusiness#Business Day#Dealbook` -1.619e+14
## myCategory.fctrmyOther -3.620e+15
## `myCategory.fctrBusiness#Technology#` -2.259e+14
## `myCategory.fctrBusiness#Crosswords/Games#` 1.208e+15
## `myCategory.fctrTStyle##` -6.498e+14
## `myCategory.fctrForeign#World#` -3.509e+15
## `myCategory.fctrOpEd#Opinion#` 9.960e+14
## `myCategory.fctrStyles##Fashion` -2.063e+15
## `myCategory.fctr#Opinion#Room For Debate` -8.404e+14
## `myCategory.fctr#U.S.#Education` -3.710e+15
## `myCategory.fctr##` -2.618e+14
## `myCategory.fctrMetro#N.Y. / Region#` 3.407e+13
## `myCategory.fctrBusiness#Business Day#Small Business` -1.683e+15
## `myCategory.fctrStyles#U.S.#` 1.780e+15
## `myCategory.fctrTravel#Travel#` -2.973e+15
## `myCategory.fctr#Opinion#The Public Editor` 6.220e+14
## A.T.one -7.820e+16
## S.T.one 7.790e+16
## A.T.state -7.949e+16
## S.T.state 7.978e+16
## H.T.take 9.187e+13
## A.npnct17.log 1.369e+14
## S.npnct17.log NA
## A.T.presid 3.136e+16
## S.T.presid -3.091e+16
## S.npnct08.log 2.586e+15
## H.T.time 1.632e+14
## A.npnct08.log NA
## S.npnct09.log -2.451e+15
## A.npnct09.log NA
## PubDate.last100.log -7.791e+12
## .rnorm -1.807e+13
## H.npnct05.log -5.209e+15
## H.T.obama -1.727e+14
## H.T.say 2.733e+13
## H.T.bank 1.426e+14
## `PubDate.date.fctr(7,13]` -2.843e+13
## `PubDate.date.fctr(13,19]` -6.481e+12
## `PubDate.date.fctr(19,25]` -4.281e+13
## `PubDate.date.fctr(25,31]` 5.777e+12
## `PubDate.second.fctr(14.8,29.5]` 2.444e+13
## `PubDate.second.fctr(29.5,44.2]` 1.108e+12
## `PubDate.second.fctr(44.2,59.1]` -9.179e+13
## H.npnct07.log 1.546e+14
## A.npnct07.log -4.672e+15
## S.npnct07.log NA
## S.npnct03.log -2.205e+15
## A.npnct19.log 9.718e+15
## H.npnct13.log 1.274e+14
## A.has.http 1.357e+16
## A.npnct03.log NA
## H.T.big -1.153e+14
## A.npnct02.log -4.296e+15
## A.npnct18.log 1.155e+16
## A.npnct20.log NA
## A.has.year.colon 3.374e+14
## S.has.year.colon NA
## A.npnct22.log -1.697e+15
## S.npnct22.log NA
## H.npnct02.log -1.092e+15
## H.T.test 1.233e+14
## S.npnct15.log 1.825e+16
## S.T.take -2.056e+17
## A.T.take 2.059e+17
## A.npnct06.log -4.292e+14
## S.npnct06.log NA
## A.npnct15.log -1.862e+16
## H.npnct14.log 7.406e+12
## H.T.deal -3.294e+15
## S.T.new -1.924e+17
## A.T.new 1.924e+17
## H.T.billion 4.103e+14
## H.T.polit -2.066e+14
## H.T.china -8.335e+13
## H.T.art -4.173e+14
## `PubDate.minute.fctr(14.8,29.5]` -1.815e+13
## `PubDate.minute.fctr(29.5,44.2]` -2.612e+13
## `PubDate.minute.fctr(44.2,59.1]` 1.046e+13
## S.npnct13.log -4.039e+15
## A.npnct13.log 3.990e+15
## A.T.will 4.084e+15
## S.T.will -4.320e+15
## PubDate.wkday.fctr1 -2.088e+14
## PubDate.wkday.fctr2 -4.149e+14
## PubDate.wkday.fctr3 -2.993e+14
## PubDate.wkday.fctr4 -3.624e+14
## PubDate.wkday.fctr5 -2.624e+14
## PubDate.wkday.fctr6 -4.528e+14
## H.T.pictur -7.310e+13
## S.T.day -8.934e+16
## S.T.show 4.759e+17
## A.T.show -4.777e+17
## A.T.day 8.954e+16
## H.T.new -3.294e+14
## S.npnct30.log -7.031e+15
## A.npnct30.log 5.431e+15
## H.T.news -1.074e+14
## H.T.first -4.305e+14
## H.T.X2014 -6.448e+14
## A.T.first -1.001e+17
## S.T.first 9.928e+16
## A.T.year 1.008e+16
## S.T.year -1.054e+16
## A.T.report 4.988e+16
## S.T.report -5.163e+16
## A.T.compani -6.701e+16
## S.T.compani 6.669e+16
## H.T.morn 7.508e+14
## H.T.busi -9.852e+14
## A.npnct14.log -3.038e+14
## A.T.share -8.452e+14
## S.T.share NA
## H.npnct04.log -9.395e+14
## S.T.time -2.987e+17
## A.T.time 2.987e+17
## S.npnct14.log 5.650e+14
## A.T.articl -3.664e+16
## S.T.articl 3.608e+16
## H.T.newyork 8.287e+13
## A.T.newyork -1.447e+17
## S.T.newyork 1.454e+17
## H.T.today -4.028e+14
## H.T.springsumm 2.518e+15
## H.T.day -6.361e+13
## H.npnct15.log -2.180e+15
## H.T.report -6.027e+14
## A.npnct04.log -2.989e+14
## S.npnct04.log NA
## H.T.daili -1.410e+15
## H.T.X2015 -2.670e+15
## S.npnct16.log 1.844e+14
## H.T.week -3.754e+14
## A.npnct16.log NA
## A.T.intern 3.504e+16
## S.T.intern -3.522e+16
## H.has.year.colon -6.216e+13
## H.T.fashion -1.950e+14
## H.npnct16.log -2.016e+14
## A.T.fashion -2.259e+17
## S.T.fashion 2.242e+17
## A.T.week -2.078e+16
## S.T.week 2.042e+16
## H.npnct30.log -1.706e+14
## S.npnct12.log 1.035e+16
## A.npnct12.log -1.038e+16
## H.ndgts.log 2.458e+14
## S.ndgts.log 1.018e+14
## A.ndgts.log -2.768e+14
## H.nuppr.log 3.490e+14
## H.nchrs.log -2.391e+14
## H.nwrds.unq.log -2.224e+14
## A.nchrs.log 1.275e+16
## S.nchrs.log -1.276e+16
## A.nwrds.unq.log -1.401e+16
## S.nwrds.unq.log 1.384e+16
## S.nuppr.log 4.663e+14
## A.nuppr.log -6.663e+14
## Std. Error
## (Intercept) 4.884e+07
## WordCount.log 1.247e+06
## H.nwrds.log 1.142e+07
## S.nwrds.log 7.539e+08
## `PubDate.hour.fctr(7.67,15.3]` 3.838e+06
## `PubDate.hour.fctr(15.3,23]` 4.062e+06
## A.nwrds.log 7.538e+08
## H.npnct21.log 6.701e+06
## PubDate.wkend 7.790e+06
## S.npnct21.log 2.699e+08
## A.npnct21.log 2.693e+08
## H.npnct08.log 1.495e+07
## H.npnct09.log NA
## PubDate.last10.log 1.963e+06
## PubDate.last1.log 7.126e+05
## A.T.make 3.334e+09
## S.T.make 3.330e+09
## H.npnct06.log 1.674e+07
## A.T.can 5.151e+08
## A.npnct01.log 2.687e+07
## S.npnct01.log NA
## S.T.can 5.152e+08
## H.npnct17.log 1.789e+07
## H.has.ebola 8.938e+06
## H.npnct01.log 2.342e+07
## A.T.said 2.378e+09
## S.T.said 2.377e+09
## H.T.make 7.632e+06
## H.npnct12.log 3.689e+06
## `myCategory.fctrForeign#World#Asia Pacific` 9.333e+06
## `myCategory.fctr#Multimedia#` 1.071e+07
## `myCategory.fctrCulture#Arts#` 7.396e+06
## `myCategory.fctrBusiness#Business Day#Dealbook` 7.293e+06
## myCategory.fctrmyOther 1.510e+07
## `myCategory.fctrBusiness#Technology#` 7.891e+06
## `myCategory.fctrBusiness#Crosswords/Games#` 1.052e+07
## `myCategory.fctrTStyle##` 7.358e+06
## `myCategory.fctrForeign#World#` 1.764e+07
## `myCategory.fctrOpEd#Opinion#` 7.401e+06
## `myCategory.fctrStyles##Fashion` 1.146e+07
## `myCategory.fctr#Opinion#Room For Debate` 1.187e+07
## `myCategory.fctr#U.S.#Education` 1.149e+07
## `myCategory.fctr##` 6.883e+06
## `myCategory.fctrMetro#N.Y. / Region#` 9.768e+06
## `myCategory.fctrBusiness#Business Day#Small Business` 1.088e+07
## `myCategory.fctrStyles#U.S.#` 9.001e+06
## `myCategory.fctrTravel#Travel#` 9.846e+06
## `myCategory.fctr#Opinion#The Public Editor` 2.277e+07
## A.T.one 3.013e+09
## S.T.one 3.011e+09
## A.T.state 2.088e+09
## S.T.state 2.085e+09
## H.T.take 7.953e+06
## A.npnct17.log 2.112e+07
## S.npnct17.log NA
## A.T.presid 9.589e+08
## S.T.presid 9.583e+08
## S.npnct08.log 5.209e+07
## H.T.time 5.071e+06
## A.npnct08.log NA
## S.npnct09.log 4.870e+07
## A.npnct09.log NA
## PubDate.last100.log 7.806e+05
## .rnorm 1.033e+06
## H.npnct05.log 4.050e+07
## H.T.obama 7.976e+06
## H.T.say 7.556e+06
## H.T.bank 9.619e+06
## `PubDate.date.fctr(7,13]` 3.247e+06
## `PubDate.date.fctr(13,19]` 3.197e+06
## `PubDate.date.fctr(19,25]` 3.097e+06
## `PubDate.date.fctr(25,31]` 3.449e+06
## `PubDate.second.fctr(14.8,29.5]` 2.871e+06
## `PubDate.second.fctr(29.5,44.2]` 2.840e+06
## `PubDate.second.fctr(44.2,59.1]` 2.903e+06
## H.npnct07.log 3.004e+06
## A.npnct07.log 5.150e+07
## S.npnct07.log NA
## S.npnct03.log 4.403e+07
## A.npnct19.log 4.966e+09
## H.npnct13.log 5.317e+06
## A.has.http 3.743e+09
## A.npnct03.log NA
## H.T.big 7.606e+06
## A.npnct02.log 6.396e+07
## A.npnct18.log 6.170e+08
## A.npnct20.log NA
## A.has.year.colon 2.392e+07
## S.has.year.colon NA
## A.npnct22.log 3.264e+07
## S.npnct22.log NA
## H.npnct02.log 2.535e+07
## H.T.test 8.862e+06
## S.npnct15.log 9.418e+08
## S.T.take 3.876e+09
## A.T.take 3.881e+09
## A.npnct06.log 1.801e+07
## S.npnct06.log NA
## A.npnct15.log 9.418e+08
## H.npnct14.log 3.473e+06
## H.T.deal 1.076e+07
## S.T.new 2.851e+09
## A.T.new 2.855e+09
## H.T.billion 1.439e+07
## H.T.polit 6.366e+06
## H.T.china 9.605e+06
## H.T.art 7.476e+06
## `PubDate.minute.fctr(14.8,29.5]` 2.954e+06
## `PubDate.minute.fctr(29.5,44.2]` 2.808e+06
## `PubDate.minute.fctr(44.2,59.1]` 3.018e+06
## S.npnct13.log 9.158e+07
## A.npnct13.log 9.143e+07
## A.T.will 6.034e+08
## S.T.will 6.024e+08
## PubDate.wkday.fctr1 9.482e+06
## PubDate.wkday.fctr2 1.010e+07
## PubDate.wkday.fctr3 1.004e+07
## PubDate.wkday.fctr4 9.893e+06
## PubDate.wkday.fctr5 1.002e+07
## PubDate.wkday.fctr6 7.850e+06
## H.T.pictur 7.985e+06
## S.T.day 1.989e+09
## S.T.show 8.052e+09
## A.T.show 8.052e+09
## A.T.day 1.993e+09
## H.T.new 6.611e+06
## S.npnct30.log 3.521e+08
## A.npnct30.log 3.522e+08
## H.T.news 9.494e+06
## H.T.first 7.207e+06
## H.T.X2014 9.811e+06
## A.T.first 2.839e+10
## S.T.first 2.839e+10
## A.T.year 3.810e+09
## S.T.year 3.809e+09
## A.T.report 9.480e+09
## S.T.report 9.480e+09
## A.T.compani 1.508e+09
## S.T.compani 1.507e+09
## H.T.morn 1.755e+07
## H.T.busi 9.082e+06
## A.npnct14.log 3.588e+07
## A.T.share 1.428e+07
## S.T.share NA
## H.npnct04.log 1.522e+07
## S.T.time 6.891e+09
## A.T.time 6.904e+09
## S.npnct14.log 3.573e+07
## A.T.articl 1.213e+10
## S.T.articl 1.213e+10
## H.T.newyork 7.555e+06
## A.T.newyork 6.516e+09
## S.T.newyork 6.511e+09
## H.T.today 7.779e+06
## H.T.springsumm 3.454e+07
## H.T.day 5.943e+06
## H.npnct15.log 4.158e+07
## H.T.report 1.141e+07
## A.npnct04.log 8.079e+06
## S.npnct04.log NA
## H.T.daili 1.191e+07
## H.T.X2015 2.122e+07
## S.npnct16.log 8.283e+06
## H.T.week 8.477e+06
## A.npnct16.log NA
## A.T.intern 2.940e+09
## S.T.intern 2.938e+09
## H.has.year.colon 1.446e+07
## H.T.fashion 1.068e+07
## H.npnct16.log 4.840e+06
## A.T.fashion 1.147e+10
## S.T.fashion 1.147e+10
## A.T.week 3.693e+09
## S.T.week 3.693e+09
## H.npnct30.log 1.605e+07
## S.npnct12.log 2.437e+08
## A.npnct12.log 2.437e+08
## H.ndgts.log 4.175e+06
## S.ndgts.log 5.970e+07
## A.ndgts.log 5.959e+07
## H.nuppr.log 7.808e+06
## H.nchrs.log 7.787e+06
## H.nwrds.unq.log 7.955e+06
## A.nchrs.log 5.118e+08
## S.nchrs.log 5.115e+08
## A.nwrds.unq.log 5.090e+08
## S.nwrds.unq.log 5.087e+08
## S.nuppr.log 1.340e+08
## A.nuppr.log 1.341e+08
## z value Pr(>|z|)
## (Intercept) -15438098 <2e-16
## WordCount.log 255281436 <2e-16
## H.nwrds.log -3512430 <2e-16
## S.nwrds.log -3727776 <2e-16
## `PubDate.hour.fctr(7.67,15.3]` -4523741 <2e-16
## `PubDate.hour.fctr(15.3,23]` -15037501 <2e-16
## A.nwrds.log 3455357 <2e-16
## H.npnct21.log 78665731 <2e-16
## PubDate.wkend -34163418 <2e-16
## S.npnct21.log 3866307 <2e-16
## A.npnct21.log -3231116 <2e-16
## H.npnct08.log 8241866 <2e-16
## H.npnct09.log NA NA
## PubDate.last10.log 36176299 <2e-16
## PubDate.last1.log -23139686 <2e-16
## A.T.make -17120056 <2e-16
## S.T.make 17042076 <2e-16
## H.npnct06.log 50883533 <2e-16
## A.T.can -2818179 <2e-16
## A.npnct01.log -3209541 <2e-16
## S.npnct01.log NA NA
## S.T.can 2968549 <2e-16
## H.npnct17.log -36046799 <2e-16
## H.has.ebola 13176029 <2e-16
## H.npnct01.log -36041309 <2e-16
## A.T.said 34802996 <2e-16
## S.T.said -34721646 <2e-16
## H.T.make 2492948 <2e-16
## H.npnct12.log 46182701 <2e-16
## `myCategory.fctrForeign#World#Asia Pacific` -323367070 <2e-16
## `myCategory.fctr#Multimedia#` -39946231 <2e-16
## `myCategory.fctrCulture#Arts#` -38235958 <2e-16
## `myCategory.fctrBusiness#Business Day#Dealbook` -22197646 <2e-16
## myCategory.fctrmyOther -239727942 <2e-16
## `myCategory.fctrBusiness#Technology#` -28627754 <2e-16
## `myCategory.fctrBusiness#Crosswords/Games#` 114875175 <2e-16
## `myCategory.fctrTStyle##` -88304590 <2e-16
## `myCategory.fctrForeign#World#` -198983262 <2e-16
## `myCategory.fctrOpEd#Opinion#` 134574469 <2e-16
## `myCategory.fctrStyles##Fashion` -179980218 <2e-16
## `myCategory.fctr#Opinion#Room For Debate` -70803818 <2e-16
## `myCategory.fctr#U.S.#Education` -322807986 <2e-16
## `myCategory.fctr##` -38037713 <2e-16
## `myCategory.fctrMetro#N.Y. / Region#` 3488166 <2e-16
## `myCategory.fctrBusiness#Business Day#Small Business` -154600943 <2e-16
## `myCategory.fctrStyles#U.S.#` 197755499 <2e-16
## `myCategory.fctrTravel#Travel#` -301962536 <2e-16
## `myCategory.fctr#Opinion#The Public Editor` 27317445 <2e-16
## A.T.one -25948697 <2e-16
## S.T.one 25874812 <2e-16
## A.T.state -38065120 <2e-16
## S.T.state 38256400 <2e-16
## H.T.take 11551544 <2e-16
## A.npnct17.log 6483162 <2e-16
## S.npnct17.log NA NA
## A.T.presid 32709030 <2e-16
## S.T.presid -32254974 <2e-16
## S.npnct08.log 49641396 <2e-16
## H.T.time 32175785 <2e-16
## A.npnct08.log NA NA
## S.npnct09.log -50331583 <2e-16
## A.npnct09.log NA NA
## PubDate.last100.log -9979834 <2e-16
## .rnorm -17495048 <2e-16
## H.npnct05.log -128597524 <2e-16
## H.T.obama -21657498 <2e-16
## H.T.say 3616440 <2e-16
## H.T.bank 14823366 <2e-16
## `PubDate.date.fctr(7,13]` -8758494 <2e-16
## `PubDate.date.fctr(13,19]` -2027034 <2e-16
## `PubDate.date.fctr(19,25]` -13822825 <2e-16
## `PubDate.date.fctr(25,31]` 1675144 <2e-16
## `PubDate.second.fctr(14.8,29.5]` 8513683 <2e-16
## `PubDate.second.fctr(29.5,44.2]` 390229 <2e-16
## `PubDate.second.fctr(44.2,59.1]` -31625335 <2e-16
## H.npnct07.log 51460041 <2e-16
## A.npnct07.log -90714148 <2e-16
## S.npnct07.log NA NA
## S.npnct03.log -50092682 <2e-16
## A.npnct19.log 1956751 <2e-16
## H.npnct13.log 23954511 <2e-16
## A.has.http 3624328 <2e-16
## A.npnct03.log NA NA
## H.T.big -15158005 <2e-16
## A.npnct02.log -67165063 <2e-16
## A.npnct18.log 18712240 <2e-16
## A.npnct20.log NA NA
## A.has.year.colon 14106914 <2e-16
## S.has.year.colon NA NA
## A.npnct22.log -51984796 <2e-16
## S.npnct22.log NA NA
## H.npnct02.log -43058995 <2e-16
## H.T.test 13914702 <2e-16
## S.npnct15.log 19374779 <2e-16
## S.T.take -53040842 <2e-16
## A.T.take 53055093 <2e-16
## A.npnct06.log -23837483 <2e-16
## S.npnct06.log NA NA
## A.npnct15.log -19775289 <2e-16
## H.npnct14.log 2132482 <2e-16
## H.T.deal -306234030 <2e-16
## S.T.new -67493238 <2e-16
## A.T.new 67408959 <2e-16
## H.T.billion 28511002 <2e-16
## H.T.polit -32457361 <2e-16
## H.T.china -8678147 <2e-16
## H.T.art -55816354 <2e-16
## `PubDate.minute.fctr(14.8,29.5]` -6145463 <2e-16
## `PubDate.minute.fctr(29.5,44.2]` -9301057 <2e-16
## `PubDate.minute.fctr(44.2,59.1]` 3465731 <2e-16
## S.npnct13.log -44101860 <2e-16
## A.npnct13.log 43636484 <2e-16
## A.T.will 6766976 <2e-16
## S.T.will -7170931 <2e-16
## PubDate.wkday.fctr1 -22023805 <2e-16
## PubDate.wkday.fctr2 -41085160 <2e-16
## PubDate.wkday.fctr3 -29807751 <2e-16
## PubDate.wkday.fctr4 -36626252 <2e-16
## PubDate.wkday.fctr5 -26193482 <2e-16
## PubDate.wkday.fctr6 -57684538 <2e-16
## H.T.pictur -9153892 <2e-16
## S.T.day -44915552 <2e-16
## S.T.show 59104424 <2e-16
## A.T.show -59325805 <2e-16
## A.T.day 44918035 <2e-16
## H.T.new -49818185 <2e-16
## S.npnct30.log -19970913 <2e-16
## A.npnct30.log 15418929 <2e-16
## H.T.news -11313054 <2e-16
## H.T.first -59726493 <2e-16
## H.T.X2014 -65723841 <2e-16
## A.T.first -3524381 <2e-16
## S.T.first 3496697 <2e-16
## A.T.year 2645424 <2e-16
## S.T.year -2766211 <2e-16
## A.T.report 5262009 <2e-16
## S.T.report -5446632 <2e-16
## A.T.compani -44444692 <2e-16
## S.T.compani 44250732 <2e-16
## H.T.morn 42771807 <2e-16
## H.T.busi -108485372 <2e-16
## A.npnct14.log -8468919 <2e-16
## A.T.share -59202645 <2e-16
## S.T.share NA NA
## H.npnct04.log -61710021 <2e-16
## S.T.time -43342196 <2e-16
## A.T.time 43270408 <2e-16
## S.npnct14.log 15809858 <2e-16
## A.T.articl -3019830 <2e-16
## S.T.articl 2974284 <2e-16
## H.T.newyork 10969016 <2e-16
## A.T.newyork -22204323 <2e-16
## S.T.newyork 22335744 <2e-16
## H.T.today -51773100 <2e-16
## H.T.springsumm 72907609 <2e-16
## H.T.day -10704599 <2e-16
## H.npnct15.log -52426398 <2e-16
## H.T.report -52802529 <2e-16
## A.npnct04.log -37000144 <2e-16
## S.npnct04.log NA NA
## H.T.daili -118352383 <2e-16
## H.T.X2015 -125805536 <2e-16
## S.npnct16.log 22266101 <2e-16
## H.T.week -44284216 <2e-16
## A.npnct16.log NA NA
## A.T.intern 11919330 <2e-16
## S.T.intern -11988721 <2e-16
## H.has.year.colon -4297536 <2e-16
## H.T.fashion -18253476 <2e-16
## H.npnct16.log -41657243 <2e-16
## A.T.fashion -19693259 <2e-16
## S.T.fashion 19544274 <2e-16
## A.T.week -5625747 <2e-16
## S.T.week 5530125 <2e-16
## H.npnct30.log -10624152 <2e-16
## S.npnct12.log 42481823 <2e-16
## A.npnct12.log -42598334 <2e-16
## H.ndgts.log 58878998 <2e-16
## S.ndgts.log 1705279 <2e-16
## A.ndgts.log -4645731 <2e-16
## H.nuppr.log 44697157 <2e-16
## H.nchrs.log -30705717 <2e-16
## H.nwrds.unq.log -27964830 <2e-16
## A.nchrs.log 24916691 <2e-16
## S.nchrs.log -24944016 <2e-16
## A.nwrds.unq.log -27535425 <2e-16
## S.nwrds.unq.log 27209404 <2e-16
## S.nuppr.log 3478858 <2e-16
## A.nuppr.log -4970273 <2e-16
##
## (Intercept) ***
## WordCount.log ***
## H.nwrds.log ***
## S.nwrds.log ***
## `PubDate.hour.fctr(7.67,15.3]` ***
## `PubDate.hour.fctr(15.3,23]` ***
## A.nwrds.log ***
## H.npnct21.log ***
## PubDate.wkend ***
## S.npnct21.log ***
## A.npnct21.log ***
## H.npnct08.log ***
## H.npnct09.log
## PubDate.last10.log ***
## PubDate.last1.log ***
## A.T.make ***
## S.T.make ***
## H.npnct06.log ***
## A.T.can ***
## A.npnct01.log ***
## S.npnct01.log
## S.T.can ***
## H.npnct17.log ***
## H.has.ebola ***
## H.npnct01.log ***
## A.T.said ***
## S.T.said ***
## H.T.make ***
## H.npnct12.log ***
## `myCategory.fctrForeign#World#Asia Pacific` ***
## `myCategory.fctr#Multimedia#` ***
## `myCategory.fctrCulture#Arts#` ***
## `myCategory.fctrBusiness#Business Day#Dealbook` ***
## myCategory.fctrmyOther ***
## `myCategory.fctrBusiness#Technology#` ***
## `myCategory.fctrBusiness#Crosswords/Games#` ***
## `myCategory.fctrTStyle##` ***
## `myCategory.fctrForeign#World#` ***
## `myCategory.fctrOpEd#Opinion#` ***
## `myCategory.fctrStyles##Fashion` ***
## `myCategory.fctr#Opinion#Room For Debate` ***
## `myCategory.fctr#U.S.#Education` ***
## `myCategory.fctr##` ***
## `myCategory.fctrMetro#N.Y. / Region#` ***
## `myCategory.fctrBusiness#Business Day#Small Business` ***
## `myCategory.fctrStyles#U.S.#` ***
## `myCategory.fctrTravel#Travel#` ***
## `myCategory.fctr#Opinion#The Public Editor` ***
## A.T.one ***
## S.T.one ***
## A.T.state ***
## S.T.state ***
## H.T.take ***
## A.npnct17.log ***
## S.npnct17.log
## A.T.presid ***
## S.T.presid ***
## S.npnct08.log ***
## H.T.time ***
## A.npnct08.log
## S.npnct09.log ***
## A.npnct09.log
## PubDate.last100.log ***
## .rnorm ***
## H.npnct05.log ***
## H.T.obama ***
## H.T.say ***
## H.T.bank ***
## `PubDate.date.fctr(7,13]` ***
## `PubDate.date.fctr(13,19]` ***
## `PubDate.date.fctr(19,25]` ***
## `PubDate.date.fctr(25,31]` ***
## `PubDate.second.fctr(14.8,29.5]` ***
## `PubDate.second.fctr(29.5,44.2]` ***
## `PubDate.second.fctr(44.2,59.1]` ***
## H.npnct07.log ***
## A.npnct07.log ***
## S.npnct07.log
## S.npnct03.log ***
## A.npnct19.log ***
## H.npnct13.log ***
## A.has.http ***
## A.npnct03.log
## H.T.big ***
## A.npnct02.log ***
## A.npnct18.log ***
## A.npnct20.log
## A.has.year.colon ***
## S.has.year.colon
## A.npnct22.log ***
## S.npnct22.log
## H.npnct02.log ***
## H.T.test ***
## S.npnct15.log ***
## S.T.take ***
## A.T.take ***
## A.npnct06.log ***
## S.npnct06.log
## A.npnct15.log ***
## H.npnct14.log ***
## H.T.deal ***
## S.T.new ***
## A.T.new ***
## H.T.billion ***
## H.T.polit ***
## H.T.china ***
## H.T.art ***
## `PubDate.minute.fctr(14.8,29.5]` ***
## `PubDate.minute.fctr(29.5,44.2]` ***
## `PubDate.minute.fctr(44.2,59.1]` ***
## S.npnct13.log ***
## A.npnct13.log ***
## A.T.will ***
## S.T.will ***
## PubDate.wkday.fctr1 ***
## PubDate.wkday.fctr2 ***
## PubDate.wkday.fctr3 ***
## PubDate.wkday.fctr4 ***
## PubDate.wkday.fctr5 ***
## PubDate.wkday.fctr6 ***
## H.T.pictur ***
## S.T.day ***
## S.T.show ***
## A.T.show ***
## A.T.day ***
## H.T.new ***
## S.npnct30.log ***
## A.npnct30.log ***
## H.T.news ***
## H.T.first ***
## H.T.X2014 ***
## A.T.first ***
## S.T.first ***
## A.T.year ***
## S.T.year ***
## A.T.report ***
## S.T.report ***
## A.T.compani ***
## S.T.compani ***
## H.T.morn ***
## H.T.busi ***
## A.npnct14.log ***
## A.T.share ***
## S.T.share
## H.npnct04.log ***
## S.T.time ***
## A.T.time ***
## S.npnct14.log ***
## A.T.articl ***
## S.T.articl ***
## H.T.newyork ***
## A.T.newyork ***
## S.T.newyork ***
## H.T.today ***
## H.T.springsumm ***
## H.T.day ***
## H.npnct15.log ***
## H.T.report ***
## A.npnct04.log ***
## S.npnct04.log
## H.T.daili ***
## H.T.X2015 ***
## S.npnct16.log ***
## H.T.week ***
## A.npnct16.log
## A.T.intern ***
## S.T.intern ***
## H.has.year.colon ***
## H.T.fashion ***
## H.npnct16.log ***
## A.T.fashion ***
## S.T.fashion ***
## A.T.week ***
## S.T.week ***
## H.npnct30.log ***
## S.npnct12.log ***
## A.npnct12.log ***
## H.ndgts.log ***
## S.ndgts.log ***
## A.ndgts.log ***
## H.nuppr.log ***
## H.nchrs.log ***
## H.nwrds.unq.log ***
## A.nchrs.log ***
## S.nchrs.log ***
## A.nwrds.unq.log ***
## S.nwrds.unq.log ***
## S.nuppr.log ***
## A.nuppr.log ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 4042.7 on 4474 degrees of freedom
## Residual deviance: 29339.5 on 4300 degrees of freedom
## AIC: 29690
##
## Number of Fisher Scoring iterations: 25
##
## [1] " calling mypredict_mdl for fit:"## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## threshold f.score
## 1 0.0 0.2867534
## 2 0.1 0.7040000
## 3 0.2 0.7040000
## 4 0.3 0.7040000
## 5 0.4 0.7040000
## 6 0.5 0.7040000
## 7 0.6 0.7040000
## 8 0.7 0.7040000
## 9 0.8 0.7040000
## 10 0.9 0.7040000
## 11 1.0 0.0000000## [1] "Classifier Probability Threshold: 0.9000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.All.X.glm.N
## 1 N 3584
## 2 Y 265
## Popular.fctr.predict.All.X.glm.Y
## 1 142
## 2 484
## Prediction
## Reference N Y
## N 3584 142
## Y 265 484
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 9.090503e-01 6.507779e-01 9.002448e-01 9.173176e-01 8.326257e-01
## AccuracyPValue McnemarPValue
## 1.896445e-49 1.472792e-09
## [1] " calling mypredict_mdl for OOB:"## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## threshold f.score
## 1 0.0 0.2865473
## 2 0.1 0.6570973
## 3 0.2 0.6570973
## 4 0.3 0.6570973
## 5 0.4 0.6570973
## 6 0.5 0.6570973
## 7 0.6 0.6570973
## 8 0.7 0.6570973
## 9 0.8 0.6570973
## 10 0.9 0.6570973
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.9000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.All.X.glm.N
## 1 N 1636
## 2 Y 138
## Popular.fctr.predict.All.X.glm.Y
## 1 77
## 2 206
## Prediction
## Reference N Y
## N 1636 77
## Y 138 206
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.954789e-01 5.961271e-01 8.814466e-01 9.083722e-01 8.327662e-01
## AccuracyPValue McnemarPValue
## 4.319079e-16 4.277313e-05
## model_id model_method
## 1 All.X.glm glm
## feats
## 1 WordCount.log, H.nwrds.log, S.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, S.npnct21.log, A.npnct21.log, H.npnct08.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, A.T.make, S.T.make, H.npnct06.log, A.T.can, A.npnct01.log, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, A.T.said, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, A.T.one, S.T.one, A.T.state, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, .rnorm, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, A.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, A.has.http, A.npnct03.log, H.T.big, A.npnct02.log, A.npnct18.log, A.npnct20.log, A.has.year.colon, S.has.year.colon, A.npnct22.log, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, A.T.take, A.npnct06.log, S.npnct06.log, A.npnct15.log, H.npnct14.log, H.T.deal, S.T.new, A.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.npnct13.log, A.T.will, S.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, A.T.show, A.T.day, H.T.new, S.npnct30.log, A.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, S.T.first, A.T.year, S.T.year, A.T.report, S.T.report, A.T.compani, S.T.compani, H.T.morn, H.T.busi, A.npnct14.log, A.T.share, S.T.share, H.npnct04.log, S.T.time, A.T.time, S.npnct14.log, A.T.articl, S.T.articl, H.T.newyork, A.T.newyork, S.T.newyork, H.T.today, H.T.springsumm, H.T.day, H.npnct15.log, H.T.report, A.npnct04.log, S.npnct04.log, H.T.daili, H.T.X2015, S.npnct16.log, H.T.week, A.npnct16.log, A.T.intern, S.T.intern, H.has.year.colon, H.T.fashion, H.npnct16.log, A.T.fashion, S.T.fashion, A.T.week, S.T.week, H.npnct30.log, S.npnct12.log, A.npnct12.log, H.ndgts.log, S.ndgts.log, A.ndgts.log, H.nuppr.log, H.nchrs.log, H.nwrds.unq.log, A.nchrs.log, S.nchrs.log, A.nwrds.unq.log, S.nwrds.unq.log, S.nuppr.log, A.nuppr.log
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 14.579 7.611
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8040422 0.9 0.704 0.8851434
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.9002448 0.9173176 0.5889746 0.7769434
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.9 0.6570973 0.8954789
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.8814466 0.9083722 0.5961271 29689.53
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.03114896 0.08080253
## label step_major step_minor bgn end elapsed
## 2 fit.models_1_glm 2 0 279.831 298.839 19.008
## 3 fit.models_1_rpart 3 0 298.839 NA NA
## [1] "fitting model: All.X.no.rnorm.rpart"
## [1] " indep_vars: WordCount.log, H.nwrds.log, S.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, S.npnct21.log, A.npnct21.log, H.npnct08.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, A.T.make, S.T.make, H.npnct06.log, A.T.can, A.npnct01.log, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, A.T.said, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, A.T.one, S.T.one, A.T.state, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, A.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, A.has.http, A.npnct03.log, H.T.big, A.npnct02.log, A.npnct18.log, A.npnct20.log, A.has.year.colon, S.has.year.colon, A.npnct22.log, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, A.T.take, A.npnct06.log, S.npnct06.log, A.npnct15.log, H.npnct14.log, H.T.deal, S.T.new, A.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.npnct13.log, A.T.will, S.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, A.T.show, A.T.day, H.T.new, S.npnct30.log, A.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, S.T.first, A.T.year, S.T.year, A.T.report, S.T.report, A.T.compani, S.T.compani, H.T.morn, H.T.busi, A.npnct14.log, A.T.share, S.T.share, H.npnct04.log, S.T.time, A.T.time, S.npnct14.log, A.T.articl, S.T.articl, H.T.newyork, A.T.newyork, S.T.newyork, H.T.today, H.T.springsumm, H.T.day, H.npnct15.log, H.T.report, A.npnct04.log, S.npnct04.log, H.T.daili, H.T.X2015, S.npnct16.log, H.T.week, A.npnct16.log, A.T.intern, S.T.intern, H.has.year.colon, H.T.fashion, H.npnct16.log, A.T.fashion, S.T.fashion, A.T.week, S.T.week, H.npnct30.log, S.npnct12.log, A.npnct12.log, H.ndgts.log, S.ndgts.log, A.ndgts.log, H.nuppr.log, H.nchrs.log, H.nwrds.unq.log, A.nchrs.log, S.nchrs.log, A.nwrds.unq.log, S.nwrds.unq.log, S.nuppr.log, A.nuppr.log"
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 0.0113 on full training set## Warning in myfit_mdl(model_id = model_id, model_method = method,
## indep_vars_vctr = indep_vars_vctr, : model's bestTune found at an extreme
## of tuneGrid for parameter: cp
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 4475
##
## CP nsplit rel error
## 1 0.27102804 0 1.0000000
## 2 0.08411215 1 0.7289720
## 3 0.01134846 2 0.6448598
##
## Variable importance
## myCategory.fctrOpEd#Opinion#
## 54
## myCategory.fctrBusiness#Crosswords/Games#
## 18
## A.nwrds.unq.log
## 7
## S.nwrds.unq.log
## 7
## A.nchrs.log
## 6
## S.nchrs.log
## 6
## H.nchrs.log
## 1
## S.nwrds.log
## 1
##
## Node number 1: 4475 observations, complexity param=0.271028
## predicted class=N expected loss=0.1673743 P(node) =1
## class counts: 3726 749
## probabilities: 0.833 0.167
## left son=2 (4106 obs) right son=3 (369 obs)
## Primary splits:
## myCategory.fctrOpEd#Opinion# < 0.5 to the left, improve=297.02950, (0 missing)
## WordCount.log < 6.524296 to the left, improve=105.72630, (0 missing)
## S.nuppr.log < 1.497866 to the right, improve= 86.35796, (0 missing)
## A.nuppr.log < 1.497866 to the right, improve= 86.35796, (0 missing)
## myCategory.fctrBusiness#Crosswords/Games# < 0.5 to the left, improve= 85.77765, (0 missing)
## Surrogate splits:
## A.nwrds.unq.log < 1.497866 to the right, agree=0.928, adj=0.127, (0 split)
## S.nwrds.unq.log < 1.497866 to the right, agree=0.928, adj=0.125, (0 split)
## A.nchrs.log < 3.725621 to the right, agree=0.927, adj=0.117, (0 split)
## S.nchrs.log < 3.725621 to the right, agree=0.927, adj=0.117, (0 split)
## S.nwrds.log < 2.564001 to the left, agree=0.918, adj=0.011, (0 split)
##
## Node number 2: 4106 observations, complexity param=0.08411215
## predicted class=N expected loss=0.1127618 P(node) =0.9175419
## class counts: 3643 463
## probabilities: 0.887 0.113
## left son=4 (4023 obs) right son=5 (83 obs)
## Primary splits:
## myCategory.fctrBusiness#Crosswords/Games# < 0.5 to the left, improve=99.60741, (0 missing)
## WordCount.log < 6.485398 to the left, improve=94.68604, (0 missing)
## myCategory.fctrStyles#U.S.# < 0.5 to the left, improve=50.94648, (0 missing)
## S.nuppr.log < 1.497866 to the right, improve=31.44556, (0 missing)
## A.nuppr.log < 1.497866 to the right, improve=31.44556, (0 missing)
## Surrogate splits:
## H.nchrs.log < 2.35024 to the right, agree=0.981, adj=0.060, (0 split)
## H.nuppr.log < 0.8958797 to the right, agree=0.980, adj=0.024, (0 split)
##
## Node number 3: 369 observations
## predicted class=Y expected loss=0.2249322 P(node) =0.0824581
## class counts: 83 286
## probabilities: 0.225 0.775
##
## Node number 4: 4023 observations
## predicted class=N expected loss=0.09694258 P(node) =0.8989944
## class counts: 3633 390
## probabilities: 0.903 0.097
##
## Node number 5: 83 observations
## predicted class=Y expected loss=0.1204819 P(node) =0.01854749
## class counts: 10 73
## probabilities: 0.120 0.880
##
## n= 4475
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 4475 749 N (0.83262570 0.16737430)
## 2) myCategory.fctrOpEd#Opinion#< 0.5 4106 463 N (0.88723819 0.11276181)
## 4) myCategory.fctrBusiness#Crosswords/Games#< 0.5 4023 390 N (0.90305742 0.09694258) *
## 5) myCategory.fctrBusiness#Crosswords/Games#>=0.5 83 10 Y (0.12048193 0.87951807) *
## 3) myCategory.fctrOpEd#Opinion#>=0.5 369 83 Y (0.22493225 0.77506775) *
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.2867534
## 2 0.1 0.5978351
## 3 0.2 0.5978351
## 4 0.3 0.5978351
## 5 0.4 0.5978351
## 6 0.5 0.5978351
## 7 0.6 0.5978351
## 8 0.7 0.5978351
## 9 0.8 0.1754808
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.7000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.All.X.no.rnorm.rpart.N
## 1 N 3633
## 2 Y 390
## Popular.fctr.predict.All.X.no.rnorm.rpart.Y
## 1 93
## 2 359
## Prediction
## Reference N Y
## N 3633 93
## Y 390 359
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.920670e-01 5.398657e-01 8.826068e-01 9.010121e-01 8.326257e-01
## AccuracyPValue McnemarPValue
## 1.439953e-29 2.397951e-41
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.2865473
## 2 0.1 0.5650558
## 3 0.2 0.5650558
## 4 0.3 0.5650558
## 5 0.4 0.5650558
## 6 0.5 0.5650558
## 7 0.6 0.5650558
## 8 0.7 0.5650558
## 9 0.8 0.1562500
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.7000 to maximize f.score.OOB"
## Popular.fctr Popular.fctr.predict.All.X.no.rnorm.rpart.N
## 1 N 1671
## 2 Y 192
## Popular.fctr.predict.All.X.no.rnorm.rpart.Y
## 1 42
## 2 152
## Prediction
## Reference N Y
## N 1671 42
## Y 192 152
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.862421e-01 5.054039e-01 8.717239e-01 8.996488e-01 8.327662e-01
## AccuracyPValue McnemarPValue
## 5.783557e-12 2.026854e-22
## model_id model_method
## 1 All.X.no.rnorm.rpart rpart
## feats
## 1 WordCount.log, H.nwrds.log, S.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, S.npnct21.log, A.npnct21.log, H.npnct08.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, A.T.make, S.T.make, H.npnct06.log, A.T.can, A.npnct01.log, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, A.T.said, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, A.T.one, S.T.one, A.T.state, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, A.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, A.has.http, A.npnct03.log, H.T.big, A.npnct02.log, A.npnct18.log, A.npnct20.log, A.has.year.colon, S.has.year.colon, A.npnct22.log, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, A.T.take, A.npnct06.log, S.npnct06.log, A.npnct15.log, H.npnct14.log, H.T.deal, S.T.new, A.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.npnct13.log, A.T.will, S.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, A.T.show, A.T.day, H.T.new, S.npnct30.log, A.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, S.T.first, A.T.year, S.T.year, A.T.report, S.T.report, A.T.compani, S.T.compani, H.T.morn, H.T.busi, A.npnct14.log, A.T.share, S.T.share, H.npnct04.log, S.T.time, A.T.time, S.npnct14.log, A.T.articl, S.T.articl, H.T.newyork, A.T.newyork, S.T.newyork, H.T.today, H.T.springsumm, H.T.day, H.npnct15.log, H.T.report, A.npnct04.log, S.npnct04.log, H.T.daili, H.T.X2015, S.npnct16.log, H.T.week, A.npnct16.log, A.T.intern, S.T.intern, H.has.year.colon, H.T.fashion, H.npnct16.log, A.T.fashion, S.T.fashion, A.T.week, S.T.week, H.npnct30.log, S.npnct12.log, A.npnct12.log, H.ndgts.log, S.ndgts.log, A.ndgts.log, H.nuppr.log, H.nchrs.log, H.nwrds.unq.log, A.nchrs.log, S.nchrs.log, A.nwrds.unq.log, S.nwrds.unq.log, S.nuppr.log, A.nuppr.log
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 3 9.482 2.102
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.7277461 0.7 0.5978351 0.8934084
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8826068 0.9010121 0.5566659 0.7084504
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.7 0.5650558 0.8862421
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8717239 0.8996488 0.5054039
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.003041136 0.02922293# User specified
# easier to exclude features
#model_id_pfx <- "";
# indep_vars_vctr <- setdiff(names(glb_fitent_df),
# union(union(glb_rsp_var, glb_exclude_vars_as_features),
# c("<feat1_name>", "<feat2_name>")))
# method <- ""
# easier to include features
#model_id_pfx <- ""; indep_vars_vctr <- c("<feat1_name>", "<feat1_name>"); method <- ""
# User specified bivariate models
# indep_vars_vctr_lst <- list()
# for (feat in setdiff(names(glb_fitent_df),
# union(glb_rsp_var, glb_exclude_vars_as_features)))
# indep_vars_vctr_lst[["feat"]] <- feat
# User specified combinatorial models
# indep_vars_vctr_lst <- list()
# combn_mtrx <- combn(c("<feat1_name>", "<feat2_name>", "<featn_name>"),
# <num_feats_to_choose>)
# for (combn_ix in 1:ncol(combn_mtrx))
# #print(combn_mtrx[, combn_ix])
# indep_vars_vctr_lst[[combn_ix]] <- combn_mtrx[, combn_ix]
# template for myfit_mdl
# rf is hard-coded in caret to recognize only Accuracy / Kappa evaluation metrics
# only for OOB in trainControl ?
# ret_lst <- myfit_mdl_fn(model_id=paste0(model_id_pfx, ""), model_method=method,
# indep_vars_vctr=indep_vars_vctr,
# rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
# fit_df=glb_fitent_df, OOB_df=glb_OOBent_df,
# n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df,
# model_loss_mtrx=glb_model_metric_terms,
# model_summaryFunction=glb_model_metric_smmry,
# model_metric=glb_model_metric,
# model_metric_maximize=glb_model_metric_maximize)
# Simplify a model
# fit_df <- glb_fitent_df; glb_mdl <- step(<complex>_mdl)
# Non-caret models
# rpart_area_mdl <- rpart(reformulate("Area", response=glb_rsp_var),
# data=glb_fitent_df, #method="class",
# control=rpart.control(cp=0.12),
# parms=list(loss=glb_model_metric_terms))
# print("rpart_sel_wlm_mdl"); prp(rpart_sel_wlm_mdl)
#
print(glb_models_df)## model_id model_method
## 1 MFO.myMFO_classfr myMFO_classfr
## 2 Random.myrandom_classfr myrandom_classfr
## 3 Max.cor.Y.cv.0.rpart rpart
## 4 Max.cor.Y.cv.0.cp.0.rpart rpart
## 5 Max.cor.Y.rpart rpart
## 6 Max.cor.Y.glm glm
## 7 Interact.High.cor.Y.glm glm
## 8 Low.cor.X.glm glm
## 9 All.X.glm glm
## 10 All.X.no.rnorm.rpart rpart
## feats
## 1 .rnorm
## 2 .rnorm
## 3 A.nuppr.log
## 4 A.nuppr.log
## 5 A.nuppr.log
## 6 A.nuppr.log
## 7 A.nuppr.log, A.nuppr.log:A.nwrds.log, A.nuppr.log:A.npnct21.log, A.nuppr.log:H.npnct09.log, A.nuppr.log:S.T.make, A.nuppr.log:H.npnct17.log, A.nuppr.log:S.T.can, A.nuppr.log:S.npnct01.log, A.nuppr.log:A.npnct23.log, A.nuppr.log:S.T.said, A.nuppr.log:A.npnct25.log, A.nuppr.log:S.T.one, A.nuppr.log:S.T.state, A.nuppr.log:S.npnct07.log, A.nuppr.log:A.npnct19.log, A.nuppr.log:S.npnct03.log, A.nuppr.log:A.npnct18.log, A.nuppr.log:A.npnct20.log, A.nuppr.log:S.has.year.colon, A.nuppr.log:S.npnct22.log, A.nuppr.log:S.T.take, A.nuppr.log:S.npnct06.log, A.nuppr.log:A.npnct02.log, A.nuppr.log:S.T.new, A.nuppr.log:S.npnct13.log, A.nuppr.log:A.T.will, A.nuppr.log:S.T.show, A.nuppr.log:S.T.day, A.nuppr.log:S.npnct30.log, A.nuppr.log:A.T.first, A.nuppr.log:A.T.year, A.nuppr.log:A.T.report, A.nuppr.log:A.T.compani, A.nuppr.log:A.npnct30.log, A.nuppr.log:S.T.share, A.nuppr.log:H.T.billion, A.nuppr.log:S.T.time, A.nuppr.log:A.npnct14.log, A.nuppr.log:A.T.articl, A.nuppr.log:A.T.newyork, A.nuppr.log:H.T.polit, A.nuppr.log:H.T.springsumm, A.nuppr.log:S.npnct04.log, A.nuppr.log:H.T.report, A.nuppr.log:H.npnct15.log, A.nuppr.log:S.npnct16.log, A.nuppr.log:A.T.intern, A.nuppr.log:S.T.intern, A.nuppr.log:H.T.X2015, A.nuppr.log:A.T.week, A.nuppr.log:S.npnct12.log, A.nuppr.log:S.ndgts.log, A.nuppr.log:H.nuppr.log, A.nuppr.log:A.nchrs.log, A.nuppr.log:S.nchrs.log, A.nuppr.log:S.nuppr.log
## 8 WordCount.log, H.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, A.npnct21.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, S.T.make, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, S.T.one, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, .rnorm, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, H.T.big, A.npnct20.log, S.has.year.colon, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, S.npnct06.log, H.npnct14.log, H.T.deal, S.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, H.T.new, S.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, A.T.year, A.T.report, A.T.compani, H.T.busi, A.npnct14.log, S.T.share, S.T.time, A.T.articl, H.T.newyork, A.T.newyork, H.T.springsumm, H.T.day, H.T.report, S.npnct04.log, S.npnct16.log, H.T.week, A.T.intern, H.T.fashion, H.npnct16.log, A.T.fashion, A.T.week, H.npnct30.log, S.npnct12.log, H.ndgts.log, S.ndgts.log, H.nuppr.log, H.nchrs.log, A.nchrs.log, A.nwrds.unq.log, S.nuppr.log
## 9 WordCount.log, H.nwrds.log, S.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, S.npnct21.log, A.npnct21.log, H.npnct08.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, A.T.make, S.T.make, H.npnct06.log, A.T.can, A.npnct01.log, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, A.T.said, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, A.T.one, S.T.one, A.T.state, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, .rnorm, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, A.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, A.has.http, A.npnct03.log, H.T.big, A.npnct02.log, A.npnct18.log, A.npnct20.log, A.has.year.colon, S.has.year.colon, A.npnct22.log, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, A.T.take, A.npnct06.log, S.npnct06.log, A.npnct15.log, H.npnct14.log, H.T.deal, S.T.new, A.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.npnct13.log, A.T.will, S.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, A.T.show, A.T.day, H.T.new, S.npnct30.log, A.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, S.T.first, A.T.year, S.T.year, A.T.report, S.T.report, A.T.compani, S.T.compani, H.T.morn, H.T.busi, A.npnct14.log, A.T.share, S.T.share, H.npnct04.log, S.T.time, A.T.time, S.npnct14.log, A.T.articl, S.T.articl, H.T.newyork, A.T.newyork, S.T.newyork, H.T.today, H.T.springsumm, H.T.day, H.npnct15.log, H.T.report, A.npnct04.log, S.npnct04.log, H.T.daili, H.T.X2015, S.npnct16.log, H.T.week, A.npnct16.log, A.T.intern, S.T.intern, H.has.year.colon, H.T.fashion, H.npnct16.log, A.T.fashion, S.T.fashion, A.T.week, S.T.week, H.npnct30.log, S.npnct12.log, A.npnct12.log, H.ndgts.log, S.ndgts.log, A.ndgts.log, H.nuppr.log, H.nchrs.log, H.nwrds.unq.log, A.nchrs.log, S.nchrs.log, A.nwrds.unq.log, S.nwrds.unq.log, S.nuppr.log, A.nuppr.log
## 10 WordCount.log, H.nwrds.log, S.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, S.npnct21.log, A.npnct21.log, H.npnct08.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, A.T.make, S.T.make, H.npnct06.log, A.T.can, A.npnct01.log, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, A.T.said, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, A.T.one, S.T.one, A.T.state, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, A.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, A.has.http, A.npnct03.log, H.T.big, A.npnct02.log, A.npnct18.log, A.npnct20.log, A.has.year.colon, S.has.year.colon, A.npnct22.log, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, A.T.take, A.npnct06.log, S.npnct06.log, A.npnct15.log, H.npnct14.log, H.T.deal, S.T.new, A.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.npnct13.log, A.T.will, S.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, A.T.show, A.T.day, H.T.new, S.npnct30.log, A.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, S.T.first, A.T.year, S.T.year, A.T.report, S.T.report, A.T.compani, S.T.compani, H.T.morn, H.T.busi, A.npnct14.log, A.T.share, S.T.share, H.npnct04.log, S.T.time, A.T.time, S.npnct14.log, A.T.articl, S.T.articl, H.T.newyork, A.T.newyork, S.T.newyork, H.T.today, H.T.springsumm, H.T.day, H.npnct15.log, H.T.report, A.npnct04.log, S.npnct04.log, H.T.daili, H.T.X2015, S.npnct16.log, H.T.week, A.npnct16.log, A.T.intern, S.T.intern, H.has.year.colon, H.T.fashion, H.npnct16.log, A.T.fashion, S.T.fashion, A.T.week, S.T.week, H.npnct30.log, S.npnct12.log, A.npnct12.log, H.ndgts.log, S.ndgts.log, A.ndgts.log, H.nuppr.log, H.nchrs.log, H.nwrds.unq.log, A.nchrs.log, S.nchrs.log, A.nwrds.unq.log, S.nwrds.unq.log, S.nuppr.log, A.nuppr.log
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 0 0.666 0.003
## 2 0 0.339 0.001
## 3 0 0.650 0.057
## 4 0 0.596 0.056
## 5 1 1.274 0.056
## 6 1 1.212 0.079
## 7 1 3.104 1.108
## 8 1 7.723 3.727
## 9 1 14.579 7.611
## 10 3 9.482 2.102
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5000000 0.5 0.0000000 0.8326257
## 2 0.5007516 0.1 0.2867534 0.1673743
## 3 0.5000000 0.5 0.0000000 0.8326257
## 4 0.5000000 0.5 0.0000000 0.8326257
## 5 0.5000000 0.5 0.0000000 0.8326258
## 6 0.7073742 0.2 0.3986014 0.8324022
## 7 0.7990119 0.3 0.4778761 0.8451393
## 8 0.9494083 0.3 0.7445072 0.9061454
## 9 0.8040422 0.9 0.7040000 0.8851434
## 10 0.7277461 0.7 0.5978351 0.8934084
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8213602 0.8434553 0.0000000000 0.5000000
## 2 0.1565447 0.1786398 0.0000000000 0.4909227
## 3 0.8213602 0.8434553 0.0000000000 0.5000000
## 4 0.8213602 0.8434553 0.0000000000 0.5000000
## 5 0.8213602 0.8434553 0.0000000000 0.5000000
## 6 0.7176970 0.7439004 -0.0004459345 0.7102060
## 7 0.7902571 0.8138164 0.1744464709 0.7758607
## 8 0.9002448 0.9173176 0.6418070528 0.9160473
## 9 0.9002448 0.9173176 0.5889745502 0.7769434
## 10 0.8826068 0.9010121 0.5566658958 0.7084504
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0.0000000 0.8327662
## 2 0.1 0.2865473 0.1672338
## 3 0.5 0.0000000 0.8327662
## 4 0.5 0.0000000 0.8327662
## 5 0.5 0.0000000 0.8327662
## 6 0.2 0.3880266 0.7316480
## 7 0.3 0.4595635 0.7953330
## 8 0.3 0.7340720 0.9066602
## 9 0.9 0.6570973 0.8954789
## 10 0.7 0.5650558 0.8862421
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8159247 0.8486533 0.0000000
## 2 0.1513467 0.1840753 0.0000000
## 3 0.8159247 0.8486533 0.0000000
## 4 0.8159247 0.8486533 0.0000000
## 5 0.8159247 0.8486533 0.0000000
## 6 0.7119353 0.7506985 0.2283681
## 7 0.7772394 0.8125808 0.3354449
## 8 0.8932593 0.9188884 0.6776204
## 9 0.8814466 0.9083722 0.5961271
## 10 0.8717239 0.8996488 0.5054039
## max.AccuracySD.fit max.KappaSD.fit min.aic.fit
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 0.0002791548 0.0000000000 NA
## 6 0.0000648833 0.0007723812 3714.601
## 7 0.0042028288 0.0170435147 3399.630
## 8 0.0011432814 0.0083812472 2099.443
## 9 0.0311489608 0.0808025302 29689.534
## 10 0.0030411362 0.0292229339 NAfit.models_1_chunk_df <- myadd_chunk(fit.models_1_chunk_df, "fit.models_1_end",
major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 3 fit.models_1_rpart 3 0 298.839 312.663 13.824
## 4 fit.models_1_end 4 0 312.663 NA NAglb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 11 fit.models 7 1 275.697 312.67 36.973
## 12 fit.models 7 2 312.670 NA NAif (!is.null(glb_model_metric_smmry)) {
stats_df <- glb_models_df[, "model_id", FALSE]
stats_mdl_df <- data.frame()
for (model_id in stats_df$model_id) {
stats_mdl_df <- rbind(stats_mdl_df,
mypredict_mdl(glb_models_lst[[model_id]], glb_fitent_df, glb_rsp_var,
glb_rsp_var_out, model_id, "fit",
glb_model_metric_smmry, glb_model_metric,
glb_model_metric_maximize, ret_type="stats"))
}
stats_df <- merge(stats_df, stats_mdl_df, all.x=TRUE)
stats_mdl_df <- data.frame()
for (model_id in stats_df$model_id) {
stats_mdl_df <- rbind(stats_mdl_df,
mypredict_mdl(glb_models_lst[[model_id]], glb_OOBent_df, glb_rsp_var,
glb_rsp_var_out, model_id, "OOB",
glb_model_metric_smmry, glb_model_metric,
glb_model_metric_maximize, ret_type="stats"))
}
stats_df <- merge(stats_df, stats_mdl_df, all.x=TRUE)
# tmp_models_df <- orderBy(~model_id, glb_models_df)
# rownames(tmp_models_df) <- seq(1, nrow(tmp_models_df))
# all.equal(subset(tmp_models_df[, names(stats_df)], model_id != "Random.myrandom_classfr"),
# subset(stats_df, model_id != "Random.myrandom_classfr"))
# print(subset(tmp_models_df[, names(stats_df)], model_id != "Random.myrandom_classfr")[, c("model_id", "max.Accuracy.fit")])
# print(subset(stats_df, model_id != "Random.myrandom_classfr")[, c("model_id", "max.Accuracy.fit")])
print("Merging following data into glb_models_df:")
print(stats_mrg_df <- stats_df[, c(1, grep(glb_model_metric, names(stats_df)))])
print(tmp_models_df <- orderBy(~model_id, glb_models_df[, c("model_id", grep(glb_model_metric, names(stats_df), value=TRUE))]))
tmp2_models_df <- glb_models_df[, c("model_id", setdiff(names(glb_models_df), grep(glb_model_metric, names(stats_df), value=TRUE)))]
tmp3_models_df <- merge(tmp2_models_df, stats_mrg_df, all.x=TRUE, sort=FALSE)
print(tmp3_models_df)
print(names(tmp3_models_df))
print(glb_models_df <- subset(tmp3_models_df, select=-model_id.1))
}
plt_models_df <- glb_models_df[, -grep("SD|Upper|Lower", names(glb_models_df))]
for (var in grep("^min.", names(plt_models_df), value=TRUE)) {
plt_models_df[, sub("min.", "inv.", var)] <-
#ifelse(all(is.na(tmp <- plt_models_df[, var])), NA, 1.0 / tmp)
1.0 / plt_models_df[, var]
plt_models_df <- plt_models_df[ , -grep(var, names(plt_models_df))]
}
print(plt_models_df)## model_id model_method
## 1 MFO.myMFO_classfr myMFO_classfr
## 2 Random.myrandom_classfr myrandom_classfr
## 3 Max.cor.Y.cv.0.rpart rpart
## 4 Max.cor.Y.cv.0.cp.0.rpart rpart
## 5 Max.cor.Y.rpart rpart
## 6 Max.cor.Y.glm glm
## 7 Interact.High.cor.Y.glm glm
## 8 Low.cor.X.glm glm
## 9 All.X.glm glm
## 10 All.X.no.rnorm.rpart rpart
## feats
## 1 .rnorm
## 2 .rnorm
## 3 A.nuppr.log
## 4 A.nuppr.log
## 5 A.nuppr.log
## 6 A.nuppr.log
## 7 A.nuppr.log, A.nuppr.log:A.nwrds.log, A.nuppr.log:A.npnct21.log, A.nuppr.log:H.npnct09.log, A.nuppr.log:S.T.make, A.nuppr.log:H.npnct17.log, A.nuppr.log:S.T.can, A.nuppr.log:S.npnct01.log, A.nuppr.log:A.npnct23.log, A.nuppr.log:S.T.said, A.nuppr.log:A.npnct25.log, A.nuppr.log:S.T.one, A.nuppr.log:S.T.state, A.nuppr.log:S.npnct07.log, A.nuppr.log:A.npnct19.log, A.nuppr.log:S.npnct03.log, A.nuppr.log:A.npnct18.log, A.nuppr.log:A.npnct20.log, A.nuppr.log:S.has.year.colon, A.nuppr.log:S.npnct22.log, A.nuppr.log:S.T.take, A.nuppr.log:S.npnct06.log, A.nuppr.log:A.npnct02.log, A.nuppr.log:S.T.new, A.nuppr.log:S.npnct13.log, A.nuppr.log:A.T.will, A.nuppr.log:S.T.show, A.nuppr.log:S.T.day, A.nuppr.log:S.npnct30.log, A.nuppr.log:A.T.first, A.nuppr.log:A.T.year, A.nuppr.log:A.T.report, A.nuppr.log:A.T.compani, A.nuppr.log:A.npnct30.log, A.nuppr.log:S.T.share, A.nuppr.log:H.T.billion, A.nuppr.log:S.T.time, A.nuppr.log:A.npnct14.log, A.nuppr.log:A.T.articl, A.nuppr.log:A.T.newyork, A.nuppr.log:H.T.polit, A.nuppr.log:H.T.springsumm, A.nuppr.log:S.npnct04.log, A.nuppr.log:H.T.report, A.nuppr.log:H.npnct15.log, A.nuppr.log:S.npnct16.log, A.nuppr.log:A.T.intern, A.nuppr.log:S.T.intern, A.nuppr.log:H.T.X2015, A.nuppr.log:A.T.week, A.nuppr.log:S.npnct12.log, A.nuppr.log:S.ndgts.log, A.nuppr.log:H.nuppr.log, A.nuppr.log:A.nchrs.log, A.nuppr.log:S.nchrs.log, A.nuppr.log:S.nuppr.log
## 8 WordCount.log, H.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, A.npnct21.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, S.T.make, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, S.T.one, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, .rnorm, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, H.T.big, A.npnct20.log, S.has.year.colon, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, S.npnct06.log, H.npnct14.log, H.T.deal, S.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, H.T.new, S.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, A.T.year, A.T.report, A.T.compani, H.T.busi, A.npnct14.log, S.T.share, S.T.time, A.T.articl, H.T.newyork, A.T.newyork, H.T.springsumm, H.T.day, H.T.report, S.npnct04.log, S.npnct16.log, H.T.week, A.T.intern, H.T.fashion, H.npnct16.log, A.T.fashion, A.T.week, H.npnct30.log, S.npnct12.log, H.ndgts.log, S.ndgts.log, H.nuppr.log, H.nchrs.log, A.nchrs.log, A.nwrds.unq.log, S.nuppr.log
## 9 WordCount.log, H.nwrds.log, S.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, S.npnct21.log, A.npnct21.log, H.npnct08.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, A.T.make, S.T.make, H.npnct06.log, A.T.can, A.npnct01.log, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, A.T.said, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, A.T.one, S.T.one, A.T.state, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, .rnorm, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, A.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, A.has.http, A.npnct03.log, H.T.big, A.npnct02.log, A.npnct18.log, A.npnct20.log, A.has.year.colon, S.has.year.colon, A.npnct22.log, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, A.T.take, A.npnct06.log, S.npnct06.log, A.npnct15.log, H.npnct14.log, H.T.deal, S.T.new, A.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.npnct13.log, A.T.will, S.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, A.T.show, A.T.day, H.T.new, S.npnct30.log, A.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, S.T.first, A.T.year, S.T.year, A.T.report, S.T.report, A.T.compani, S.T.compani, H.T.morn, H.T.busi, A.npnct14.log, A.T.share, S.T.share, H.npnct04.log, S.T.time, A.T.time, S.npnct14.log, A.T.articl, S.T.articl, H.T.newyork, A.T.newyork, S.T.newyork, H.T.today, H.T.springsumm, H.T.day, H.npnct15.log, H.T.report, A.npnct04.log, S.npnct04.log, H.T.daili, H.T.X2015, S.npnct16.log, H.T.week, A.npnct16.log, A.T.intern, S.T.intern, H.has.year.colon, H.T.fashion, H.npnct16.log, A.T.fashion, S.T.fashion, A.T.week, S.T.week, H.npnct30.log, S.npnct12.log, A.npnct12.log, H.ndgts.log, S.ndgts.log, A.ndgts.log, H.nuppr.log, H.nchrs.log, H.nwrds.unq.log, A.nchrs.log, S.nchrs.log, A.nwrds.unq.log, S.nwrds.unq.log, S.nuppr.log, A.nuppr.log
## 10 WordCount.log, H.nwrds.log, S.nwrds.log, PubDate.hour.fctr, A.nwrds.log, H.npnct21.log, PubDate.wkend, S.npnct21.log, A.npnct21.log, H.npnct08.log, H.npnct09.log, PubDate.last10.log, PubDate.last1.log, A.T.make, S.T.make, H.npnct06.log, A.T.can, A.npnct01.log, S.npnct01.log, S.T.can, H.npnct17.log, H.has.ebola, H.npnct01.log, A.T.said, S.T.said, H.T.make, H.npnct12.log, myCategory.fctr, A.T.one, S.T.one, A.T.state, S.T.state, H.T.take, A.npnct17.log, S.npnct17.log, A.T.presid, S.T.presid, S.npnct08.log, H.T.time, A.npnct08.log, S.npnct09.log, A.npnct09.log, PubDate.last100.log, H.npnct05.log, H.T.obama, H.T.say, H.T.bank, PubDate.date.fctr, PubDate.second.fctr, H.npnct07.log, A.npnct07.log, S.npnct07.log, S.npnct03.log, A.npnct19.log, H.npnct13.log, A.has.http, A.npnct03.log, H.T.big, A.npnct02.log, A.npnct18.log, A.npnct20.log, A.has.year.colon, S.has.year.colon, A.npnct22.log, S.npnct22.log, H.npnct02.log, H.T.test, S.npnct15.log, S.T.take, A.T.take, A.npnct06.log, S.npnct06.log, A.npnct15.log, H.npnct14.log, H.T.deal, S.T.new, A.T.new, H.T.billion, H.T.polit, H.T.china, H.T.art, PubDate.minute.fctr, S.npnct13.log, A.npnct13.log, A.T.will, S.T.will, PubDate.wkday.fctr, H.T.pictur, S.T.day, S.T.show, A.T.show, A.T.day, H.T.new, S.npnct30.log, A.npnct30.log, H.T.news, H.T.first, H.T.X2014, A.T.first, S.T.first, A.T.year, S.T.year, A.T.report, S.T.report, A.T.compani, S.T.compani, H.T.morn, H.T.busi, A.npnct14.log, A.T.share, S.T.share, H.npnct04.log, S.T.time, A.T.time, S.npnct14.log, A.T.articl, S.T.articl, H.T.newyork, A.T.newyork, S.T.newyork, H.T.today, H.T.springsumm, H.T.day, H.npnct15.log, H.T.report, A.npnct04.log, S.npnct04.log, H.T.daili, H.T.X2015, S.npnct16.log, H.T.week, A.npnct16.log, A.T.intern, S.T.intern, H.has.year.colon, H.T.fashion, H.npnct16.log, A.T.fashion, S.T.fashion, A.T.week, S.T.week, H.npnct30.log, S.npnct12.log, A.npnct12.log, H.ndgts.log, S.ndgts.log, A.ndgts.log, H.nuppr.log, H.nchrs.log, H.nwrds.unq.log, A.nchrs.log, S.nchrs.log, A.nwrds.unq.log, S.nwrds.unq.log, S.nuppr.log, A.nuppr.log
## max.nTuningRuns max.auc.fit opt.prob.threshold.fit max.f.score.fit
## 1 0 0.5000000 0.5 0.0000000
## 2 0 0.5007516 0.1 0.2867534
## 3 0 0.5000000 0.5 0.0000000
## 4 0 0.5000000 0.5 0.0000000
## 5 1 0.5000000 0.5 0.0000000
## 6 1 0.7073742 0.2 0.3986014
## 7 1 0.7990119 0.3 0.4778761
## 8 1 0.9494083 0.3 0.7445072
## 9 1 0.8040422 0.9 0.7040000
## 10 3 0.7277461 0.7 0.5978351
## max.Accuracy.fit max.Kappa.fit max.auc.OOB opt.prob.threshold.OOB
## 1 0.8326257 0.0000000000 0.5000000 0.5
## 2 0.1673743 0.0000000000 0.4909227 0.1
## 3 0.8326257 0.0000000000 0.5000000 0.5
## 4 0.8326257 0.0000000000 0.5000000 0.5
## 5 0.8326258 0.0000000000 0.5000000 0.5
## 6 0.8324022 -0.0004459345 0.7102060 0.2
## 7 0.8451393 0.1744464709 0.7758607 0.3
## 8 0.9061454 0.6418070528 0.9160473 0.3
## 9 0.8851434 0.5889745502 0.7769434 0.9
## 10 0.8934084 0.5566658958 0.7084504 0.7
## max.f.score.OOB max.Accuracy.OOB max.Kappa.OOB
## 1 0.0000000 0.8327662 0.0000000
## 2 0.2865473 0.1672338 0.0000000
## 3 0.0000000 0.8327662 0.0000000
## 4 0.0000000 0.8327662 0.0000000
## 5 0.0000000 0.8327662 0.0000000
## 6 0.3880266 0.7316480 0.2283681
## 7 0.4595635 0.7953330 0.3354449
## 8 0.7340720 0.9066602 0.6776204
## 9 0.6570973 0.8954789 0.5961271
## 10 0.5650558 0.8862421 0.5054039
## inv.elapsedtime.everything inv.elapsedtime.final inv.aic.fit
## 1 1.50150150 333.3333333 NA
## 2 2.94985251 1000.0000000 NA
## 3 1.53846154 17.5438596 NA
## 4 1.67785235 17.8571429 NA
## 5 0.78492936 17.8571429 NA
## 6 0.82508251 12.6582278 0.0002692079
## 7 0.32216495 0.9025271 0.0002941496
## 8 0.12948336 0.2683123 0.0004763168
## 9 0.06859181 0.1313888 0.0000336819
## 10 0.10546298 0.4757374 NAprint(myplot_radar(radar_inp_df=plt_models_df))## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 10. Consider specifying shapes manually. if you must have them.## Warning: Removed 5 rows containing missing values (geom_path).## Warning: Removed 60 rows containing missing values (geom_point).## Warning: Removed 6 rows containing missing values (geom_text).## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 10. Consider specifying shapes manually. if you must have them.
# print(myplot_radar(radar_inp_df=subset(plt_models_df,
# !(model_id %in% grep("random|MFO", plt_models_df$model_id, value=TRUE)))))
# Compute CI for <metric>SD
glb_models_df <- mutate(glb_models_df,
max.df = ifelse(max.nTuningRuns > 1, max.nTuningRuns - 1, NA),
min.sd2ci.scaler = ifelse(is.na(max.df), NA, qt(0.975, max.df)))
for (var in grep("SD", names(glb_models_df), value=TRUE)) {
# Does CI alredy exist ?
var_components <- unlist(strsplit(var, "SD"))
varActul <- paste0(var_components[1], var_components[2])
varUpper <- paste0(var_components[1], "Upper", var_components[2])
varLower <- paste0(var_components[1], "Lower", var_components[2])
if (varUpper %in% names(glb_models_df)) {
warning(varUpper, " already exists in glb_models_df")
# Assuming Lower also exists
next
}
print(sprintf("var:%s", var))
# CI is dependent on sample size in t distribution; df=n-1
glb_models_df[, varUpper] <- glb_models_df[, varActul] +
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
glb_models_df[, varLower] <- glb_models_df[, varActul] -
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
}## Warning: max.AccuracyUpper.fit already exists in glb_models_df## [1] "var:max.KappaSD.fit"# Plot metrics with CI
plt_models_df <- glb_models_df[, "model_id", FALSE]
pltCI_models_df <- glb_models_df[, "model_id", FALSE]
for (var in grep("Upper", names(glb_models_df), value=TRUE)) {
var_components <- unlist(strsplit(var, "Upper"))
col_name <- unlist(paste(var_components, collapse=""))
plt_models_df[, col_name] <- glb_models_df[, col_name]
for (name in paste0(var_components[1], c("Upper", "Lower"), var_components[2]))
pltCI_models_df[, name] <- glb_models_df[, name]
}
build_statsCI_data <- function(plt_models_df) {
mltd_models_df <- melt(plt_models_df, id.vars="model_id")
mltd_models_df$data <- sapply(1:nrow(mltd_models_df),
function(row_ix) tail(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]), "[.]")), 1))
mltd_models_df$label <- sapply(1:nrow(mltd_models_df),
function(row_ix) head(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]),
paste0(".", mltd_models_df[row_ix, "data"]))), 1))
#print(mltd_models_df)
return(mltd_models_df)
}
mltd_models_df <- build_statsCI_data(plt_models_df)
mltdCI_models_df <- melt(pltCI_models_df, id.vars="model_id")
for (row_ix in 1:nrow(mltdCI_models_df)) {
for (type in c("Upper", "Lower")) {
if (length(var_components <- unlist(strsplit(
as.character(mltdCI_models_df[row_ix, "variable"]), type))) > 1) {
#print(sprintf("row_ix:%d; type:%s; ", row_ix, type))
mltdCI_models_df[row_ix, "label"] <- var_components[1]
mltdCI_models_df[row_ix, "data"] <-
unlist(strsplit(var_components[2], "[.]"))[2]
mltdCI_models_df[row_ix, "type"] <- type
break
}
}
}
#print(mltdCI_models_df)
# castCI_models_df <- dcast(mltdCI_models_df, value ~ type, fun.aggregate=sum)
# print(castCI_models_df)
wideCI_models_df <- reshape(subset(mltdCI_models_df, select=-variable),
timevar="type",
idvar=setdiff(names(mltdCI_models_df), c("type", "value", "variable")),
direction="wide")
#print(wideCI_models_df)
mrgdCI_models_df <- merge(wideCI_models_df, mltd_models_df, all.x=TRUE)
#print(mrgdCI_models_df)
# Merge stats back in if CIs don't exist
goback_vars <- c()
for (var in unique(mltd_models_df$label)) {
for (type in unique(mltd_models_df$data)) {
var_type <- paste0(var, ".", type)
# if this data is already present, next
if (var_type %in% unique(paste(mltd_models_df$label, mltd_models_df$data,
sep=".")))
next
#print(sprintf("var_type:%s", var_type))
goback_vars <- c(goback_vars, var_type)
}
}
if (length(goback_vars) > 0) {
mltd_goback_df <- build_statsCI_data(glb_models_df[, c("model_id", goback_vars)])
mltd_models_df <- rbind(mltd_models_df, mltd_goback_df)
}
mltd_models_df <- merge(mltd_models_df, glb_models_df[, c("model_id", "model_method")],
all.x=TRUE)
png(paste0(glb_out_pfx, "models_bar.png"), width=480*3, height=480*2)
print(gp <- myplot_bar(mltd_models_df, "model_id", "value", colorcol_name="model_method") +
geom_errorbar(data=mrgdCI_models_df,
mapping=aes(x=model_id, ymax=value.Upper, ymin=value.Lower), width=0.5) +
facet_grid(label ~ data, scales="free") +
theme(axis.text.x = element_text(angle = 90,vjust = 0.5)))## Warning: Stacking not well defined when ymin != 0dev.off()## quartz_off_screen
## 2print(gp)## Warning: Stacking not well defined when ymin != 0
# used for console inspection
model_evl_terms <- c(NULL)
for (metric in glb_model_evl_criteria)
model_evl_terms <- c(model_evl_terms,
ifelse(length(grep("max", metric)) > 0, "-", "+"), metric)
if (glb_is_classification && glb_is_binomial)
model_evl_terms <- c(model_evl_terms, "-", "opt.prob.threshold.OOB")
model_sel_frmla <- as.formula(paste(c("~ ", model_evl_terms), collapse=" "))
print(dsp_models_df <- orderBy(model_sel_frmla, glb_models_df)
[, c("model_id", glb_model_evl_criteria,
ifelse(glb_is_classification && glb_is_binomial,
"opt.prob.threshold.OOB", NULL))])## model_id max.Accuracy.OOB max.auc.OOB max.Kappa.OOB
## 8 Low.cor.X.glm 0.9066602 0.9160473 0.6776204
## 9 All.X.glm 0.8954789 0.7769434 0.5961271
## 10 All.X.no.rnorm.rpart 0.8862421 0.7084504 0.5054039
## 1 MFO.myMFO_classfr 0.8327662 0.5000000 0.0000000
## 3 Max.cor.Y.cv.0.rpart 0.8327662 0.5000000 0.0000000
## 4 Max.cor.Y.cv.0.cp.0.rpart 0.8327662 0.5000000 0.0000000
## 5 Max.cor.Y.rpart 0.8327662 0.5000000 0.0000000
## 7 Interact.High.cor.Y.glm 0.7953330 0.7758607 0.3354449
## 6 Max.cor.Y.glm 0.7316480 0.7102060 0.2283681
## 2 Random.myrandom_classfr 0.1672338 0.4909227 0.0000000
## min.aic.fit opt.prob.threshold.OOB
## 8 2099.443 0.3
## 9 29689.534 0.9
## 10 NA 0.7
## 1 NA 0.5
## 3 NA 0.5
## 4 NA 0.5
## 5 NA 0.5
## 7 3399.630 0.3
## 6 3714.601 0.2
## 2 NA 0.1print(myplot_radar(radar_inp_df=dsp_models_df))## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 10. Consider specifying shapes manually. if you must have them.## Warning: Removed 27 rows containing missing values (geom_point).## Warning: Removed 6 rows containing missing values (geom_text).## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 10. Consider specifying shapes manually. if you must have them.
print("Metrics used for model selection:"); print(model_sel_frmla)## [1] "Metrics used for model selection:"## ~-max.Accuracy.OOB - max.auc.OOB - max.Kappa.OOB + min.aic.fit -
## opt.prob.threshold.OOBprint(sprintf("Best model id: %s", dsp_models_df[1, "model_id"]))## [1] "Best model id: Low.cor.X.glm"if (is.null(glb_sel_mdl_id)) {
glb_sel_mdl_id <- dsp_models_df[1, "model_id"]
if (glb_sel_mdl_id == "Interact.High.cor.Y.glm") {
warning("glb_sel_mdl_id: Interact.High.cor.Y.glm; myextract_mdl_feats does not currently support interaction terms")
glb_sel_mdl_id <- dsp_models_df[2, "model_id"]
}
} else
print(sprintf("User specified selection: %s", glb_sel_mdl_id))
myprint_mdl(glb_sel_mdl <- glb_models_lst[[glb_sel_mdl_id]])## Warning: not plotting observations with leverage one:
## 2501
## Warning: not plotting observations with leverage one:
## 2501
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.8204 -0.3159 -0.1312 0.0000 3.6505
##
## Coefficients: (3 not defined because of singularities)
## Estimate
## (Intercept) -2.748e+00
## WordCount.log 1.127e+00
## H.nwrds.log 5.287e-01
## `PubDate.hour.fctr(7.67,15.3]` 2.343e-01
## `PubDate.hour.fctr(15.3,23]` 4.066e-01
## A.nwrds.log -9.763e-01
## H.npnct21.log 1.488e+00
## PubDate.wkend -3.372e-01
## A.npnct21.log 1.435e+00
## H.npnct09.log 1.945e+00
## PubDate.last10.log 2.252e-01
## PubDate.last1.log -4.492e-02
## S.T.make -1.135e+00
## S.npnct01.log 2.001e+00
## S.T.can -1.274e+00
## H.npnct17.log 1.060e+00
## H.has.ebola -2.699e-01
## H.npnct01.log -1.496e+00
## S.T.said 1.473e+00
## H.T.make -1.680e-01
## H.npnct12.log 4.681e-01
## `myCategory.fctrForeign#World#Asia Pacific` -3.838e+00
## `myCategory.fctr#Multimedia#` -4.232e+00
## `myCategory.fctrCulture#Arts#` -2.823e+00
## `myCategory.fctrBusiness#Business Day#Dealbook` -2.536e+00
## myCategory.fctrmyOther -2.117e+01
## `myCategory.fctrBusiness#Technology#` -1.867e+00
## `myCategory.fctrBusiness#Crosswords/Games#` 8.317e-01
## `myCategory.fctrTStyle##` -4.344e+00
## `myCategory.fctrForeign#World#` -1.883e+01
## `myCategory.fctrOpEd#Opinion#` 6.538e-01
## `myCategory.fctrStyles##Fashion` -2.269e+01
## `myCategory.fctr#Opinion#Room For Debate` -5.785e+00
## `myCategory.fctr#U.S.#Education` -2.168e+01
## `myCategory.fctr##` -2.624e+00
## `myCategory.fctrMetro#N.Y. / Region#` -2.037e+00
## `myCategory.fctrBusiness#Business Day#Small Business` -4.395e+00
## `myCategory.fctrStyles#U.S.#` -4.788e-01
## `myCategory.fctrTravel#Travel#` -4.023e+00
## `myCategory.fctr#Opinion#The Public Editor` 1.075e+00
## S.T.one -1.037e+00
## S.T.state 1.478e+00
## H.T.take -2.639e-01
## A.npnct17.log -1.698e-01
## S.npnct17.log NA
## A.T.presid 5.367e+02
## S.T.presid -5.362e+02
## S.npnct08.log 1.379e+01
## H.T.time 1.491e-01
## A.npnct08.log NA
## S.npnct09.log -1.233e+01
## A.npnct09.log NA
## PubDate.last100.log 1.446e-02
## .rnorm -7.241e-02
## H.npnct05.log -2.532e+01
## H.T.obama -1.444e-01
## H.T.say -6.471e-01
## H.T.bank 2.742e-01
## `PubDate.date.fctr(7,13]` -6.259e-02
## `PubDate.date.fctr(13,19]` -1.762e-01
## `PubDate.date.fctr(19,25]` -1.147e-01
## `PubDate.date.fctr(25,31]` 1.184e-01
## `PubDate.second.fctr(14.8,29.5]` 1.115e-01
## `PubDate.second.fctr(29.5,44.2]` 1.025e-03
## `PubDate.second.fctr(44.2,59.1]` -2.817e-01
## H.npnct07.log 2.272e-01
## S.npnct07.log -3.359e+01
## S.npnct03.log -2.905e+01
## A.npnct19.log -2.303e+01
## H.npnct13.log 3.531e-01
## H.T.big -4.463e-01
## A.npnct20.log -3.310e+00
## S.has.year.colon -1.323e+01
## S.npnct22.log -2.488e+01
## H.npnct02.log -1.826e+01
## H.T.test -1.230e-01
## S.npnct15.log 3.617e-01
## S.T.take -5.749e-01
## S.npnct06.log 1.534e-01
## H.npnct14.log -1.969e-01
## H.T.deal -2.294e+01
## S.T.new 5.514e-02
## H.T.billion -3.929e-01
## H.T.polit -8.283e-01
## H.T.china -7.701e-01
## H.T.art -1.001e+00
## `PubDate.minute.fctr(14.8,29.5]` -1.444e-01
## `PubDate.minute.fctr(29.5,44.2]` -2.290e-01
## `PubDate.minute.fctr(44.2,59.1]` 1.875e-03
## S.npnct13.log -1.692e-01
## A.T.will -1.093e+00
## PubDate.wkday.fctr1 -6.483e-01
## PubDate.wkday.fctr2 -1.251e+00
## PubDate.wkday.fctr3 -8.655e-01
## PubDate.wkday.fctr4 -1.092e+00
## PubDate.wkday.fctr5 -9.206e-01
## PubDate.wkday.fctr6 -1.272e+00
## H.T.pictur 2.972e-01
## S.T.day -5.547e-01
## S.T.show -1.515e+00
## H.T.new -4.940e-01
## S.npnct30.log -1.513e+01
## H.T.news -5.654e-01
## H.T.first -9.638e-01
## H.T.X2014 -3.545e-01
## A.T.first -2.167e-01
## A.T.year -7.364e-01
## A.T.report -2.921e+00
## A.T.compani -9.640e-01
## H.T.busi -6.922e-01
## A.npnct14.log 8.924e-01
## S.T.share -1.700e+00
## S.T.time -1.078e+00
## A.T.articl -6.541e-01
## H.T.newyork -7.851e-01
## A.T.newyork 2.404e+00
## H.T.springsumm -1.732e+01
## H.T.day -4.617e-01
## H.T.report -1.326e+00
## S.npnct04.log -1.164e+00
## S.npnct16.log 1.580e-01
## H.T.week -3.672e-01
## A.T.intern -2.783e+00
## H.T.fashion 2.117e+00
## H.npnct16.log -4.049e-01
## A.T.fashion -5.294e+01
## A.T.week -1.209e+00
## H.npnct30.log -1.121e-01
## S.npnct12.log -1.786e-01
## H.ndgts.log 2.950e-01
## S.ndgts.log -3.106e-01
## H.nuppr.log 1.115e+00
## H.nchrs.log -1.433e+00
## A.nchrs.log 5.040e-01
## A.nwrds.unq.log -1.111e+00
## S.nuppr.log -6.250e-01
## Std. Error z value
## (Intercept) 2.988e+00 -0.920
## WordCount.log 9.057e-02 12.444
## H.nwrds.log 6.890e-01 0.767
## `PubDate.hour.fctr(7.67,15.3]` 2.454e-01 0.955
## `PubDate.hour.fctr(15.3,23]` 2.486e-01 1.636
## A.nwrds.log 8.029e-01 -1.216
## H.npnct21.log 3.166e-01 4.701
## PubDate.wkend 4.501e-01 -0.749
## A.npnct21.log 3.318e-01 4.325
## H.npnct09.log 7.141e-01 2.724
## PubDate.last10.log 1.245e-01 1.809
## PubDate.last1.log 4.431e-02 -1.014
## S.T.make 6.030e-01 -1.882
## S.npnct01.log 1.838e+00 1.089
## S.T.can 8.334e-01 -1.528
## H.npnct17.log 5.751e-01 1.843
## H.has.ebola 4.468e-01 -0.604
## H.npnct01.log 1.264e+00 -1.183
## S.T.said 7.783e-01 1.892
## H.T.make 3.354e-01 -0.501
## H.npnct12.log 2.093e-01 2.236
## `myCategory.fctrForeign#World#Asia Pacific` 6.885e-01 -5.575
## `myCategory.fctr#Multimedia#` 8.108e-01 -5.220
## `myCategory.fctrCulture#Arts#` 3.668e-01 -7.697
## `myCategory.fctrBusiness#Business Day#Dealbook` 3.157e-01 -8.034
## myCategory.fctrmyOther 3.027e+03 -0.007
## `myCategory.fctrBusiness#Technology#` 3.268e-01 -5.713
## `myCategory.fctrBusiness#Crosswords/Games#` 5.119e-01 1.625
## `myCategory.fctrTStyle##` 5.226e-01 -8.312
## `myCategory.fctrForeign#World#` 1.402e+03 -0.013
## `myCategory.fctrOpEd#Opinion#` 2.980e-01 2.194
## `myCategory.fctrStyles##Fashion` 1.487e+03 -0.015
## `myCategory.fctr#Opinion#Room For Debate` 6.357e-01 -9.099
## `myCategory.fctr#U.S.#Education` 9.806e+02 -0.022
## `myCategory.fctr##` 2.935e-01 -8.942
## `myCategory.fctrMetro#N.Y. / Region#` 4.855e-01 -4.196
## `myCategory.fctrBusiness#Business Day#Small Business` 7.101e-01 -6.189
## `myCategory.fctrStyles#U.S.#` 3.358e-01 -1.426
## `myCategory.fctrTravel#Travel#` 1.048e+00 -3.837
## `myCategory.fctr#Opinion#The Public Editor` 1.224e+00 0.878
## S.T.one 6.057e-01 -1.712
## S.T.state 8.398e-01 1.760
## H.T.take 4.617e-01 -0.572
## A.npnct17.log 1.347e+00 -0.126
## S.npnct17.log NA NA
## A.T.presid 1.572e+05 0.003
## S.T.presid 1.572e+05 -0.003
## S.npnct08.log 1.279e+04 0.001
## H.T.time 3.019e-01 0.494
## A.npnct08.log NA NA
## S.npnct09.log 1.279e+04 -0.001
## A.npnct09.log NA NA
## PubDate.last100.log 4.532e-02 0.319
## .rnorm 6.298e-02 -1.150
## H.npnct05.log 1.030e+04 -0.002
## H.T.obama 4.409e-01 -0.328
## H.T.say 4.253e-01 -1.522
## H.T.bank 4.816e-01 0.569
## `PubDate.date.fctr(7,13]` 1.968e-01 -0.318
## `PubDate.date.fctr(13,19]` 1.939e-01 -0.909
## `PubDate.date.fctr(19,25]` 1.917e-01 -0.598
## `PubDate.date.fctr(25,31]` 2.062e-01 0.574
## `PubDate.second.fctr(14.8,29.5]` 1.748e-01 0.638
## `PubDate.second.fctr(29.5,44.2]` 1.711e-01 0.006
## `PubDate.second.fctr(44.2,59.1]` 1.778e-01 -1.584
## H.npnct07.log 1.853e-01 1.226
## S.npnct07.log 9.747e+03 -0.003
## S.npnct03.log 8.794e+03 -0.003
## A.npnct19.log 4.681e+04 0.000
## H.npnct13.log 3.115e-01 1.134
## H.T.big 5.970e-01 -0.748
## A.npnct20.log 2.891e+04 0.000
## S.has.year.colon 4.978e+03 -0.003
## S.npnct22.log 7.190e+03 -0.003
## H.npnct02.log 5.027e+03 -0.004
## H.T.test 7.093e-01 -0.173
## S.npnct15.log 1.488e+00 0.243
## S.T.take 1.016e+00 -0.566
## S.npnct06.log 1.549e+00 0.099
## H.npnct14.log 1.970e-01 -1.000
## H.T.deal 2.771e+03 -0.008
## S.T.new 7.210e-01 0.076
## H.T.billion 8.425e-01 -0.466
## H.T.polit 3.243e-01 -2.554
## H.T.china 9.744e-01 -0.790
## H.T.art 8.356e-01 -1.198
## `PubDate.minute.fctr(14.8,29.5]` 1.815e-01 -0.796
## `PubDate.minute.fctr(29.5,44.2]` 1.766e-01 -1.296
## `PubDate.minute.fctr(44.2,59.1]` 1.813e-01 0.010
## S.npnct13.log 2.026e-01 -0.835
## A.T.will 7.776e-01 -1.405
## PubDate.wkday.fctr1 5.264e-01 -1.232
## PubDate.wkday.fctr2 5.741e-01 -2.179
## PubDate.wkday.fctr3 5.659e-01 -1.529
## PubDate.wkday.fctr4 5.599e-01 -1.951
## PubDate.wkday.fctr5 5.662e-01 -1.626
## PubDate.wkday.fctr6 4.718e-01 -2.695
## H.T.pictur 6.365e-01 0.467
## S.T.day 9.902e-01 -0.560
## S.T.show 1.117e+00 -1.356
## H.T.new 4.799e-01 -1.029
## S.npnct30.log 2.101e+03 -0.007
## H.T.news 8.223e-01 -0.688
## H.T.first 9.930e-01 -0.971
## H.T.X2014 9.690e-01 -0.366
## A.T.first 1.032e+00 -0.210
## A.T.year 9.576e-01 -0.769
## A.T.report 1.211e+00 -2.412
## A.T.compani 9.255e-01 -1.042
## H.T.busi 7.758e-01 -0.892
## A.npnct14.log 2.594e-01 3.440
## S.T.share 1.039e+00 -1.636
## S.T.time 9.376e-01 -1.150
## A.T.articl 1.915e+00 -0.342
## H.T.newyork 4.970e-01 -1.580
## A.T.newyork 9.848e-01 2.441
## H.T.springsumm 1.589e+03 -0.011
## H.T.day 5.816e-01 -0.794
## H.T.report 8.203e-01 -1.617
## S.npnct04.log 6.808e-01 -1.710
## S.npnct16.log 4.908e-01 0.322
## H.T.week 5.517e-01 -0.666
## A.T.intern 2.640e+00 -1.054
## H.T.fashion 1.478e+00 1.432
## H.npnct16.log 2.864e-01 -1.414
## A.T.fashion 2.505e+03 -0.021
## A.T.week 8.310e-01 -1.454
## H.npnct30.log 1.698e+00 -0.066
## S.npnct12.log 1.445e-01 -1.236
## H.ndgts.log 2.290e-01 1.288
## S.ndgts.log 1.539e-01 -2.018
## H.nuppr.log 4.205e-01 2.652
## H.nchrs.log 3.596e-01 -3.984
## A.nchrs.log 4.909e-01 1.027
## A.nwrds.unq.log 5.515e-01 -2.015
## S.nuppr.log 1.551e-01 -4.030
## Pr(>|z|)
## (Intercept) 0.357701
## WordCount.log < 2e-16 ***
## H.nwrds.log 0.442895
## `PubDate.hour.fctr(7.67,15.3]` 0.339763
## `PubDate.hour.fctr(15.3,23]` 0.101890
## A.nwrds.log 0.224007
## H.npnct21.log 2.59e-06 ***
## PubDate.wkend 0.453664
## A.npnct21.log 1.53e-05 ***
## H.npnct09.log 0.006447 **
## PubDate.last10.log 0.070513 .
## PubDate.last1.log 0.310622
## S.T.make 0.059858 .
## S.npnct01.log 0.276171
## S.T.can 0.126439
## H.npnct17.log 0.065363 .
## H.has.ebola 0.545777
## H.npnct01.log 0.236626
## S.T.said 0.058431 .
## H.T.make 0.616373
## H.npnct12.log 0.025343 *
## `myCategory.fctrForeign#World#Asia Pacific` 2.47e-08 ***
## `myCategory.fctr#Multimedia#` 1.79e-07 ***
## `myCategory.fctrCulture#Arts#` 1.39e-14 ***
## `myCategory.fctrBusiness#Business Day#Dealbook` 9.44e-16 ***
## myCategory.fctrmyOther 0.994419
## `myCategory.fctrBusiness#Technology#` 1.11e-08 ***
## `myCategory.fctrBusiness#Crosswords/Games#` 0.104210
## `myCategory.fctrTStyle##` < 2e-16 ***
## `myCategory.fctrForeign#World#` 0.989285
## `myCategory.fctrOpEd#Opinion#` 0.028231 *
## `myCategory.fctrStyles##Fashion` 0.987826
## `myCategory.fctr#Opinion#Room For Debate` < 2e-16 ***
## `myCategory.fctr#U.S.#Education` 0.982357
## `myCategory.fctr##` < 2e-16 ***
## `myCategory.fctrMetro#N.Y. / Region#` 2.72e-05 ***
## `myCategory.fctrBusiness#Business Day#Small Business` 6.04e-10 ***
## `myCategory.fctrStyles#U.S.#` 0.153955
## `myCategory.fctrTravel#Travel#` 0.000124 ***
## `myCategory.fctr#Opinion#The Public Editor` 0.380018
## S.T.one 0.086839 .
## S.T.state 0.078395 .
## H.T.take 0.567613
## A.npnct17.log 0.899679
## S.npnct17.log NA
## A.T.presid 0.997275
## S.T.presid 0.997278
## S.npnct08.log 0.999140
## H.T.time 0.621318
## A.npnct08.log NA
## S.npnct09.log 0.999231
## A.npnct09.log NA
## PubDate.last100.log 0.749746
## .rnorm 0.250250
## H.npnct05.log 0.998038
## H.T.obama 0.743216
## H.T.say 0.128070
## H.T.bank 0.569143
## `PubDate.date.fctr(7,13]` 0.750494
## `PubDate.date.fctr(13,19]` 0.363555
## `PubDate.date.fctr(19,25]` 0.549773
## `PubDate.date.fctr(25,31]` 0.565707
## `PubDate.second.fctr(14.8,29.5]` 0.523613
## `PubDate.second.fctr(29.5,44.2]` 0.995222
## `PubDate.second.fctr(44.2,59.1]` 0.113090
## H.npnct07.log 0.220030
## S.npnct07.log 0.997250
## S.npnct03.log 0.997364
## A.npnct19.log 0.999608
## H.npnct13.log 0.256931
## H.T.big 0.454674
## A.npnct20.log 0.999909
## S.has.year.colon 0.997879
## S.npnct22.log 0.997239
## H.npnct02.log 0.997102
## H.T.test 0.862283
## S.npnct15.log 0.807992
## S.T.take 0.571550
## S.npnct06.log 0.921110
## H.npnct14.log 0.317493
## H.T.deal 0.993395
## S.T.new 0.939036
## H.T.billion 0.640945
## H.T.polit 0.010639 *
## H.T.china 0.429307
## H.T.art 0.231053
## `PubDate.minute.fctr(14.8,29.5]` 0.426227
## `PubDate.minute.fctr(29.5,44.2]` 0.194882
## `PubDate.minute.fctr(44.2,59.1]` 0.991751
## S.npnct13.log 0.403614
## A.T.will 0.159909
## PubDate.wkday.fctr1 0.218114
## PubDate.wkday.fctr2 0.029324 *
## PubDate.wkday.fctr3 0.126171
## PubDate.wkday.fctr4 0.051047 .
## PubDate.wkday.fctr5 0.103932
## PubDate.wkday.fctr6 0.007038 **
## H.T.pictur 0.640490
## S.T.day 0.575321
## S.T.show 0.174976
## H.T.new 0.303333
## S.npnct30.log 0.994257
## H.T.news 0.491721
## H.T.first 0.331766
## H.T.X2014 0.714511
## A.T.first 0.833697
## A.T.year 0.441870
## A.T.report 0.015855 *
## A.T.compani 0.297572
## H.T.busi 0.372281
## A.npnct14.log 0.000581 ***
## S.T.share 0.101745
## S.T.time 0.250299
## A.T.articl 0.732637
## H.T.newyork 0.114193
## A.T.newyork 0.014657 *
## H.T.springsumm 0.991302
## H.T.day 0.427300
## H.T.report 0.105934
## S.npnct04.log 0.087299 .
## S.npnct16.log 0.747427
## H.T.week 0.505725
## A.T.intern 0.291668
## H.T.fashion 0.152005
## H.npnct16.log 0.157409
## A.T.fashion 0.983137
## A.T.week 0.145880
## H.npnct30.log 0.947362
## S.npnct12.log 0.216499
## H.ndgts.log 0.197683
## S.ndgts.log 0.043615 *
## H.nuppr.log 0.007997 **
## H.nchrs.log 6.78e-05 ***
## A.nchrs.log 0.304604
## A.nwrds.unq.log 0.043897 *
## S.nuppr.log 5.57e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 4042.7 on 4474 degrees of freedom
## Residual deviance: 1833.4 on 4342 degrees of freedom
## AIC: 2099.4
##
## Number of Fisher Scoring iterations: 19## [1] TRUE# From here to save(), this should all be in one function
# these are executed in the same seq twice more:
# fit.data.training & predict.data.new chunks
glb_get_predictions <- function(df, mdl_id, rsp_var_out, prob_threshold_def=NULL) {
mdl <- glb_models_lst[[mdl_id]]
rsp_var_out <- paste0(rsp_var_out, mdl_id)
if (glb_is_regression) {
df[, rsp_var_out] <- predict(mdl, newdata=df, type="raw")
print(myplot_scatter(df, glb_rsp_var, rsp_var_out, smooth=TRUE))
df[, paste0(rsp_var_out, ".err")] <-
abs(df[, rsp_var_out] - df[, glb_rsp_var])
print(head(orderBy(reformulate(c("-", paste0(glb_rsp_var_out, ".err"))),
df)))
}
if (glb_is_classification && glb_is_binomial) {
prob_threshold <- glb_models_df[glb_models_df$model_id == mdl_id,
"opt.prob.threshold.OOB"]
if (is.null(prob_threshold) || is.na(prob_threshold)) {
warning("Using default probability threshold: ", prob_threshold_def)
if (is.null(prob_threshold <- prob_threshold_def))
stop("Default probability threshold is NULL")
}
df[, paste0(rsp_var_out, ".prob")] <-
predict(mdl, newdata=df, type="prob")[, 2]
df[, rsp_var_out] <-
factor(levels(df[, glb_rsp_var])[
(df[, paste0(rsp_var_out, ".prob")] >=
prob_threshold) * 1 + 1], levels(df[, glb_rsp_var]))
# prediction stats already reported by myfit_mdl ???
}
if (glb_is_classification && !glb_is_binomial) {
df[, rsp_var_out] <- predict(mdl, newdata=df, type="raw")
}
return(df)
}
glb_OOBent_df <- glb_get_predictions(df=glb_OOBent_df, glb_sel_mdl_id, glb_rsp_var_out)## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleadingpredct_accurate_var_name <- paste0(glb_rsp_var_out, glb_sel_mdl_id, ".accurate")
glb_OOBent_df[, predct_accurate_var_name] <-
(glb_OOBent_df[, glb_rsp_var] ==
glb_OOBent_df[, paste0(glb_rsp_var_out, glb_sel_mdl_id)])
glb_feats_df <-
mymerge_feats_importance(feats_df=glb_feats_df, sel_mdl=glb_sel_mdl, glb_fitent_df)
glb_feats_df[, paste0(glb_sel_mdl_id, ".importance")] <- glb_feats_df$importance
print(glb_feats_df)## id cor.y exclude.as.feat
## WordCount.log WordCount.log 2.656836e-01 FALSE
## myCategory.fctr myCategory.fctr 1.234541e-02 FALSE
## H.npnct21.log H.npnct21.log 1.283641e-01 FALSE
## A.npnct21.log A.npnct21.log 5.482747e-02 FALSE
## S.nuppr.log S.nuppr.log -2.718459e-01 FALSE
## H.nchrs.log H.nchrs.log -1.710624e-01 FALSE
## A.npnct14.log A.npnct14.log -4.999563e-02 FALSE
## H.npnct09.log H.npnct09.log 5.375262e-02 FALSE
## PubDate.wkday.fctr PubDate.wkday.fctr -3.980129e-02 FALSE
## H.nuppr.log H.nuppr.log -1.278085e-01 FALSE
## H.T.polit H.T.polit -3.062866e-02 FALSE
## A.T.newyork A.T.newyork -5.706083e-02 FALSE
## A.T.report A.T.report -4.741555e-02 FALSE
## H.npnct12.log H.npnct12.log 1.333613e-02 FALSE
## S.ndgts.log S.ndgts.log -1.242046e-01 FALSE
## A.nwrds.unq.log A.nwrds.unq.log -2.506012e-01 FALSE
## S.T.said S.T.said 1.826884e-02 FALSE
## S.T.make S.T.make 3.959645e-02 FALSE
## H.npnct17.log H.npnct17.log 3.039622e-02 FALSE
## PubDate.last10.log PubDate.last10.log 4.931702e-02 FALSE
## S.T.state S.T.state 1.012205e-02 FALSE
## S.T.one S.T.one 1.080534e-02 FALSE
## S.npnct04.log S.npnct04.log -6.294642e-02 FALSE
## S.T.share S.T.share -5.070234e-02 FALSE
## PubDate.hour.fctr PubDate.hour.fctr 1.354368e-01 FALSE
## H.T.report H.T.report -6.244050e-02 FALSE
## PubDate.second.fctr PubDate.second.fctr -1.187946e-02 FALSE
## H.T.newyork H.T.newyork -5.650839e-02 FALSE
## S.T.can S.T.can 3.049697e-02 FALSE
## H.T.say H.T.say -9.960773e-03 FALSE
## A.T.week A.T.week -8.492895e-02 FALSE
## H.T.fashion H.T.fashion -7.947505e-02 FALSE
## H.npnct16.log H.npnct16.log -8.273237e-02 FALSE
## A.T.will A.T.will -3.887937e-02 FALSE
## S.T.show S.T.show -4.193803e-02 FALSE
## PubDate.minute.fctr PubDate.minute.fctr -3.407385e-02 FALSE
## H.ndgts.log H.ndgts.log -1.196633e-01 FALSE
## S.npnct12.log S.npnct12.log -9.158156e-02 FALSE
## H.npnct07.log H.npnct07.log -1.201741e-02 FALSE
## A.nwrds.log A.nwrds.log 1.354108e-01 FALSE
## H.T.art H.T.art -3.280483e-02 FALSE
## H.npnct01.log H.npnct01.log 2.271577e-02 FALSE
## .rnorm .rnorm -8.244230e-03 FALSE
## S.T.time S.T.time -5.303654e-02 FALSE
## H.npnct13.log H.npnct13.log -1.305305e-02 FALSE
## S.npnct01.log S.npnct01.log 3.093101e-02 FALSE
## A.T.intern A.T.intern -6.949870e-02 FALSE
## A.T.compani A.T.compani -4.751471e-02 FALSE
## H.T.new H.T.new -4.327803e-02 FALSE
## A.nchrs.log A.nchrs.log -2.245488e-01 FALSE
## PubDate.last1.log PubDate.last1.log 4.635751e-02 FALSE
## H.npnct14.log H.npnct14.log -2.524770e-02 FALSE
## H.T.first H.T.first -4.458885e-02 FALSE
## PubDate.date.fctr PubDate.date.fctr -1.164756e-02 FALSE
## H.T.busi H.T.busi -4.901905e-02 FALSE
## S.npnct13.log S.npnct13.log -3.638891e-02 FALSE
## H.T.day H.T.day -6.033488e-02 FALSE
## H.T.china H.T.china -3.144808e-02 FALSE
## A.T.year A.T.year -4.721236e-02 FALSE
## H.nwrds.log H.nwrds.log 1.410282e-01 FALSE
## PubDate.wkend PubDate.wkend 1.067288e-01 FALSE
## H.T.big H.T.big -1.390748e-02 FALSE
## H.T.news H.T.news -4.415284e-02 FALSE
## H.T.week H.T.week -6.812724e-02 FALSE
## H.has.ebola H.has.ebola 2.588140e-02 FALSE
## H.T.take H.T.take -8.582583e-04 FALSE
## H.T.bank H.T.bank -9.989139e-03 FALSE
## S.T.take S.T.take -2.275732e-02 FALSE
## S.T.day S.T.day -4.188671e-02 FALSE
## H.T.make H.T.make 1.430572e-02 FALSE
## H.T.time H.T.time -2.527450e-03 FALSE
## H.T.pictur H.T.pictur -3.993172e-02 FALSE
## H.T.billion H.T.billion -2.949817e-02 FALSE
## H.T.X2014 H.T.X2014 -4.497745e-02 FALSE
## A.T.articl A.T.articl -5.445243e-02 FALSE
## H.T.obama H.T.obama -9.907543e-03 FALSE
## S.npnct16.log S.npnct16.log -6.770952e-02 FALSE
## PubDate.last100.log PubDate.last100.log -7.663322e-03 FALSE
## S.npnct15.log S.npnct15.log -2.121844e-02 FALSE
## A.T.first A.T.first -4.603341e-02 FALSE
## H.T.test H.T.test -2.065255e-02 FALSE
## A.npnct17.log A.npnct17.log -1.587454e-03 FALSE
## S.npnct06.log S.npnct06.log -2.389145e-02 FALSE
## S.T.new S.T.new -2.769558e-02 FALSE
## H.npnct30.log H.npnct30.log -8.917338e-02 FALSE
## A.T.fashion A.T.fashion -8.419345e-02 FALSE
## H.T.springsumm H.T.springsumm -5.943248e-02 FALSE
## H.T.deal H.T.deal -2.559418e-02 FALSE
## S.npnct30.log S.npnct30.log -4.370037e-02 FALSE
## H.npnct02.log H.npnct02.log -2.001851e-02 FALSE
## S.npnct22.log S.npnct22.log -1.923169e-02 FALSE
## S.npnct07.log S.npnct07.log -1.214357e-02 FALSE
## A.T.presid A.T.presid -1.789086e-03 FALSE
## S.T.presid S.T.presid -2.079562e-03 FALSE
## S.npnct03.log S.npnct03.log -1.240734e-02 FALSE
## S.has.year.colon S.has.year.colon -1.755336e-02 FALSE
## H.npnct05.log H.npnct05.log -9.653967e-03 FALSE
## S.npnct08.log S.npnct08.log -2.413868e-03 FALSE
## S.npnct09.log S.npnct09.log -3.986882e-03 FALSE
## A.npnct19.log A.npnct19.log -1.271661e-02 FALSE
## A.npnct20.log A.npnct20.log -1.451467e-02 FALSE
## A.has.http A.has.http -1.359260e-02 FALSE
## A.has.year.colon A.has.year.colon -1.755336e-02 FALSE
## A.ndgts.log A.ndgts.log -1.249484e-01 FALSE
## A.npnct01.log A.npnct01.log 3.093101e-02 FALSE
## A.npnct02.log A.npnct02.log -1.451467e-02 FALSE
## A.npnct03.log A.npnct03.log -1.359260e-02 FALSE
## A.npnct04.log A.npnct04.log -6.294642e-02 FALSE
## A.npnct05.log A.npnct05.log NA FALSE
## A.npnct06.log A.npnct06.log -2.389145e-02 FALSE
## A.npnct07.log A.npnct07.log -1.214357e-02 FALSE
## A.npnct08.log A.npnct08.log -3.258100e-03 FALSE
## A.npnct09.log A.npnct09.log -4.775988e-03 FALSE
## A.npnct10.log A.npnct10.log NA FALSE
## A.npnct11.log A.npnct11.log -5.547032e-03 FALSE
## A.npnct12.log A.npnct12.log -9.183870e-02 FALSE
## A.npnct13.log A.npnct13.log -3.760012e-02 FALSE
## A.npnct15.log A.npnct15.log -2.407715e-02 FALSE
## A.npnct16.log A.npnct16.log -6.893301e-02 FALSE
## A.npnct18.log A.npnct18.log -1.451467e-02 FALSE
## A.npnct22.log A.npnct22.log -1.923169e-02 FALSE
## A.npnct23.log A.npnct23.log 1.537569e-02 FALSE
## A.npnct24.log A.npnct24.log NA FALSE
## A.npnct25.log A.npnct25.log 1.537569e-02 FALSE
## A.npnct26.log A.npnct26.log -9.890046e-19 FALSE
## A.npnct27.log A.npnct27.log -5.547032e-03 FALSE
## A.npnct28.log A.npnct28.log NA FALSE
## A.npnct29.log A.npnct29.log NA FALSE
## A.npnct30.log A.npnct30.log -4.373349e-02 FALSE
## A.npnct31.log A.npnct31.log NA FALSE
## A.npnct32.log A.npnct32.log NA FALSE
## A.nuppr.log A.nuppr.log -2.720962e-01 FALSE
## A.T.can A.T.can 3.127063e-02 FALSE
## A.T.day A.T.day -4.196599e-02 FALSE
## A.T.make A.T.make 3.965722e-02 FALSE
## A.T.new A.T.new -2.782876e-02 FALSE
## A.T.one A.T.one 1.081694e-02 FALSE
## A.T.said A.T.said 1.839871e-02 FALSE
## A.T.share A.T.share -5.070234e-02 FALSE
## A.T.show A.T.show -4.196129e-02 FALSE
## A.T.state A.T.state 1.020706e-02 FALSE
## A.T.take A.T.take -2.282555e-02 FALSE
## A.T.time A.T.time -5.313395e-02 FALSE
## clusterid clusterid NA FALSE
## H.has.http H.has.http NA FALSE
## H.has.year.colon H.has.year.colon -7.842875e-02 FALSE
## H.npnct03.log H.npnct03.log 9.533020e-03 FALSE
## H.npnct04.log H.npnct04.log -5.126277e-02 FALSE
## H.npnct06.log H.npnct06.log 3.190718e-02 FALSE
## H.npnct08.log H.npnct08.log 5.375262e-02 FALSE
## H.npnct10.log H.npnct10.log NA FALSE
## H.npnct11.log H.npnct11.log -5.547032e-03 FALSE
## H.npnct15.log H.npnct15.log -6.158577e-02 FALSE
## H.npnct18.log H.npnct18.log NA FALSE
## H.npnct19.log H.npnct19.log NA FALSE
## H.npnct20.log H.npnct20.log NA FALSE
## H.npnct22.log H.npnct22.log -5.547032e-03 FALSE
## H.npnct23.log H.npnct23.log NA FALSE
## H.npnct24.log H.npnct24.log NA FALSE
## H.npnct25.log H.npnct25.log NA FALSE
## H.npnct26.log H.npnct26.log -9.890046e-19 FALSE
## H.npnct27.log H.npnct27.log NA FALSE
## H.npnct28.log H.npnct28.log NA FALSE
## H.npnct29.log H.npnct29.log NA FALSE
## H.npnct31.log H.npnct31.log NA FALSE
## H.npnct32.log H.npnct32.log NA FALSE
## H.nwrds.unq.log H.nwrds.unq.log -2.044964e-01 FALSE
## H.T.daili H.T.daili -6.299948e-02 FALSE
## H.T.morn H.T.morn -4.838380e-02 FALSE
## H.T.today H.T.today -5.831308e-02 FALSE
## H.T.X2015 H.T.X2015 -6.601141e-02 FALSE
## Popular Popular 1.000000e+00 TRUE
## Popular.fctr Popular.fctr NA TRUE
## PubDate.last1 PubDate.last1 3.592267e-02 TRUE
## PubDate.last10 PubDate.last10 5.398093e-02 TRUE
## PubDate.last100 PubDate.last100 3.989229e-02 TRUE
## PubDate.month.fctr PubDate.month.fctr 1.914874e-02 TRUE
## PubDate.POSIX PubDate.POSIX 1.568326e-02 TRUE
## PubDate.year.fctr PubDate.year.fctr NA FALSE
## PubDate.zoo PubDate.zoo 1.568326e-02 TRUE
## S.has.http S.has.http NA FALSE
## S.nchrs.log S.nchrs.log -2.246930e-01 FALSE
## S.npnct02.log S.npnct02.log -5.547032e-03 FALSE
## S.npnct05.log S.npnct05.log NA FALSE
## S.npnct10.log S.npnct10.log NA FALSE
## S.npnct11.log S.npnct11.log -5.547032e-03 FALSE
## S.npnct14.log S.npnct14.log -5.332519e-02 FALSE
## S.npnct17.log S.npnct17.log -1.587454e-03 FALSE
## S.npnct18.log S.npnct18.log NA FALSE
## S.npnct19.log S.npnct19.log NA FALSE
## S.npnct20.log S.npnct20.log NA FALSE
## S.npnct21.log S.npnct21.log 5.503894e-02 FALSE
## S.npnct23.log S.npnct23.log 2.760321e-02 FALSE
## S.npnct24.log S.npnct24.log NA FALSE
## S.npnct25.log S.npnct25.log 2.760321e-02 FALSE
## S.npnct26.log S.npnct26.log -9.890046e-19 FALSE
## S.npnct27.log S.npnct27.log NA FALSE
## S.npnct28.log S.npnct28.log NA FALSE
## S.npnct29.log S.npnct29.log NA FALSE
## S.npnct31.log S.npnct31.log NA FALSE
## S.npnct32.log S.npnct32.log NA FALSE
## S.nwrds.log S.nwrds.log 1.359149e-01 FALSE
## S.nwrds.unq.log S.nwrds.unq.log -2.507969e-01 FALSE
## S.T.articl S.T.articl -5.446201e-02 FALSE
## S.T.compani S.T.compani -4.764341e-02 FALSE
## S.T.fashion S.T.fashion -8.419711e-02 FALSE
## S.T.first S.T.first -4.617532e-02 FALSE
## S.T.intern S.T.intern -6.953750e-02 FALSE
## S.T.newyork S.T.newyork -5.712853e-02 FALSE
## S.T.report S.T.report -4.746920e-02 FALSE
## S.T.week S.T.week -8.503373e-02 FALSE
## S.T.will S.T.will -3.892267e-02 FALSE
## S.T.year S.T.year -4.729011e-02 FALSE
## UniqueID UniqueID 1.182492e-02 TRUE
## WordCount WordCount 2.575265e-01 TRUE
## cor.y.abs cor.high.X freqRatio
## WordCount.log 2.656836e-01 <NA> 1.300000
## myCategory.fctr 1.234541e-02 <NA> 1.337185
## H.npnct21.log 1.283641e-01 <NA> 14.995098
## A.npnct21.log 5.482747e-02 <NA> 12.798715
## S.nuppr.log 2.718459e-01 <NA> 1.152620
## H.nchrs.log 1.710624e-01 <NA> 1.023810
## A.npnct14.log 4.999563e-02 <NA> 4.603330
## H.npnct09.log 5.375262e-02 <NA> 111.620690
## PubDate.wkday.fctr 3.980129e-02 <NA> 1.003268
## H.nuppr.log 1.278085e-01 <NA> 1.033930
## H.T.polit 3.062866e-02 <NA> 126.254902
## A.T.newyork 5.706083e-02 <NA> 88.724638
## A.T.report 4.741555e-02 <NA> 78.362500
## H.npnct12.log 1.333613e-02 <NA> 4.937442
## S.ndgts.log 1.242046e-01 <NA> 10.511247
## A.nwrds.unq.log 2.506012e-01 <NA> 1.061567
## S.T.said 1.826884e-02 <NA> 190.242424
## S.T.make 3.959645e-02 <NA> 273.782609
## H.npnct17.log 3.039622e-02 <NA> 96.104478
## PubDate.last10.log 4.931702e-02 <NA> 1.666667
## S.T.state 1.012205e-02 <NA> 315.750000
## S.T.one 1.080534e-02 <NA> 240.038462
## S.npnct04.log 6.294642e-02 <NA> 28.536364
## S.T.share 5.070234e-02 <NA> 218.448276
## PubDate.hour.fctr 1.354368e-01 <NA> 1.835040
## H.T.report 6.244050e-02 <NA> 102.000000
## PubDate.second.fctr 1.187946e-02 <NA> 1.018204
## H.T.newyork 5.650839e-02 <NA> 112.446429
## S.T.can 3.049697e-02 <NA> 261.666667
## H.T.say 9.960773e-03 <NA> 247.461538
## A.T.week 8.492895e-02 <NA> 57.122642
## H.T.fashion 7.947505e-02 <NA> 76.926829
## H.npnct16.log 8.273237e-02 <NA> 3.914910
## A.T.will 3.887937e-02 <NA> 114.711538
## S.T.show 4.193803e-02 <NA> 274.608696
## PubDate.minute.fctr 3.407385e-02 <NA> 1.483365
## H.ndgts.log 1.196633e-01 <NA> 13.616137
## S.npnct12.log 9.158156e-02 <NA> 1.660473
## H.npnct07.log 1.201741e-02 <NA> 5.437234
## A.nwrds.log 1.354108e-01 <NA> 2.583333
## H.T.art 3.280483e-02 <NA> 307.333333
## H.npnct01.log 2.271577e-02 <NA> 282.913043
## .rnorm 8.244230e-03 <NA> 2.000000
## S.T.time 5.303654e-02 <NA> 65.096774
## H.npnct13.log 1.305305e-02 <NA> 13.126638
## S.npnct01.log 3.093101e-02 <NA> 309.952381
## A.T.intern 6.949870e-02 <NA> 137.347826
## A.T.compani 4.751471e-02 <NA> 140.227273
## H.T.new 4.327803e-02 <NA> 116.333333
## A.nchrs.log 2.245488e-01 <NA> 1.328571
## PubDate.last1.log 4.635751e-02 <NA> 1.142857
## H.npnct14.log 2.524770e-02 <NA> 22.802326
## H.T.first 4.458885e-02 <NA> 194.727273
## PubDate.date.fctr 1.164756e-02 <NA> 1.021394
## H.T.busi 4.901905e-02 <NA> 229.428571
## S.npnct13.log 3.638891e-02 <NA> 5.706263
## H.T.day 6.033488e-02 <NA> 86.547945
## H.T.china 3.144808e-02 <NA> 238.555556
## A.T.year 4.721236e-02 <NA> 167.108108
## H.nwrds.log 1.410282e-01 <NA> 1.127273
## PubDate.wkend 1.067288e-01 <NA> 9.095827
## H.T.big 1.390748e-02 <NA> 403.562500
## H.T.news 4.415284e-02 <NA> 238.518519
## H.T.week 6.812724e-02 <NA> 64.071429
## H.has.ebola 2.588140e-02 <NA> 73.227273
## H.T.take 8.582583e-04 <NA> 306.904762
## H.T.bank 9.989139e-03 <NA> 221.689655
## S.T.take 2.275732e-02 <NA> 287.090909
## S.T.day 4.188671e-02 <NA> 89.600000
## H.T.make 1.430572e-02 <NA> 322.200000
## H.T.time 2.527450e-03 <NA> 247.538462
## H.T.pictur 3.993172e-02 <NA> 104.032258
## H.T.billion 2.949817e-02 <NA> 229.892857
## H.T.X2014 4.497745e-02 <NA> 112.824561
## A.T.articl 5.445243e-02 <NA> 85.500000
## H.T.obama 9.907543e-03 <NA> 229.750000
## S.npnct16.log 6.770952e-02 <NA> 13.647191
## PubDate.last100.log 7.663322e-03 <NA> 25.000000
## S.npnct15.log 2.121844e-02 <NA> 203.062500
## A.T.first 4.603341e-02 <NA> 203.709677
## H.T.test 2.065255e-02 <NA> 280.000000
## A.npnct17.log 1.587454e-03 <NA> 434.133333
## S.npnct06.log 2.389145e-02 <NA> 115.642857
## S.T.new 2.769558e-02 <NA> 107.872727
## H.npnct30.log 8.917338e-02 <NA> 24.123077
## A.T.fashion 8.419345e-02 <NA> 59.809524
## H.T.springsumm 5.943248e-02 <NA> 106.966667
## H.T.deal 2.559418e-02 <NA> 258.080000
## S.npnct30.log 4.370037e-02 <NA> 134.791667
## H.npnct02.log 2.001851e-02 <NA> 501.461538
## S.npnct22.log 1.923169e-02 <NA> 543.333333
## S.npnct07.log 1.214357e-02 <NA> 1631.750000
## A.T.presid 1.789086e-03 <NA> 241.692308
## S.T.presid 2.079562e-03 <NA> 241.692308
## S.npnct03.log 1.240734e-02 <NA> 1305.400000
## S.has.year.colon 1.755336e-02 <NA> 652.200000
## H.npnct05.log 9.653967e-03 <NA> 543.333333
## S.npnct08.log 2.413868e-03 <NA> 175.513514
## S.npnct09.log 3.986882e-03 <NA> 175.486486
## A.npnct19.log 1.271661e-02 <NA> 1631.500000
## A.npnct20.log 1.451467e-02 <NA> 1087.500000
## A.has.http 1.359260e-02 A.npnct19.log 1087.666667
## A.has.year.colon 1.755336e-02 S.has.year.colon 652.200000
## A.ndgts.log 1.249484e-01 S.ndgts.log 10.501022
## A.npnct01.log 3.093101e-02 S.npnct01.log 309.952381
## A.npnct02.log 1.451467e-02 A.npnct18.log 1087.500000
## A.npnct03.log 1.359260e-02 S.npnct03.log 1087.666667
## A.npnct04.log 6.294642e-02 S.npnct04.log 28.536364
## A.npnct05.log NA <NA> 0.000000
## A.npnct06.log 2.389145e-02 S.npnct06.log 115.642857
## A.npnct07.log 1.214357e-02 S.npnct07.log 1631.750000
## A.npnct08.log 3.258100e-03 <NA> 170.868421
## A.npnct09.log 4.775988e-03 <NA> 170.842105
## A.npnct10.log NA <NA> 0.000000
## A.npnct11.log 5.547032e-03 <NA> 6531.000000
## A.npnct12.log 9.183870e-02 S.npnct12.log 1.660473
## A.npnct13.log 3.760012e-02 S.npnct13.log 5.715368
## A.npnct15.log 2.407715e-02 A.npnct02.log 196.696970
## A.npnct16.log 6.893301e-02 S.npnct16.log 13.482222
## A.npnct18.log 1.451467e-02 A.npnct20.log 1087.500000
## A.npnct22.log 1.923169e-02 S.npnct22.log 543.333333
## A.npnct23.log 1.537569e-02 A.npnct25.log 3264.500000
## A.npnct24.log NA <NA> 0.000000
## A.npnct25.log 1.537569e-02 <NA> 3264.500000
## A.npnct26.log 9.890046e-19 <NA> 0.000000
## A.npnct27.log 5.547032e-03 <NA> 6531.000000
## A.npnct28.log NA <NA> 0.000000
## A.npnct29.log NA <NA> 0.000000
## A.npnct30.log 4.373349e-02 S.npnct30.log 126.862745
## A.npnct31.log NA <NA> 0.000000
## A.npnct32.log NA <NA> 0.000000
## A.nuppr.log 2.720962e-01 S.nuppr.log 1.151308
## A.T.can 3.127063e-02 S.T.can 261.666667
## A.T.day 4.196599e-02 S.T.day 89.585714
## A.T.make 3.965722e-02 S.T.make 273.782609
## A.T.new 2.782876e-02 S.T.new 107.836364
## A.T.one 1.081694e-02 S.T.one 240.000000
## A.T.said 1.839871e-02 S.T.said 190.242424
## A.T.share 5.070234e-02 S.T.share 218.448276
## A.T.show 4.196129e-02 S.T.show 274.608696
## A.T.state 1.020706e-02 S.T.state 315.700000
## A.T.take 2.282555e-02 S.T.take 287.045455
## A.T.time 5.313395e-02 S.T.time 65.086022
## clusterid NA <NA> 0.000000
## H.has.http NA <NA> 0.000000
## H.has.year.colon 7.842875e-02 S.T.intern 32.670103
## H.npnct03.log 9.533020e-03 <NA> 2176.333333
## H.npnct04.log 5.126277e-02 H.T.billion 38.325301
## H.npnct06.log 3.190718e-02 H.npnct17.log 68.935484
## H.npnct08.log 5.375262e-02 H.npnct09.log 111.620690
## H.npnct10.log NA <NA> 0.000000
## H.npnct11.log 5.547032e-03 <NA> 6531.000000
## H.npnct15.log 6.158577e-02 H.T.springsumm 52.983471
## H.npnct18.log NA <NA> 0.000000
## H.npnct19.log NA <NA> 0.000000
## H.npnct20.log NA <NA> 0.000000
## H.npnct22.log 5.547032e-03 <NA> 6531.000000
## H.npnct23.log NA <NA> 0.000000
## H.npnct24.log NA <NA> 0.000000
## H.npnct25.log NA <NA> 0.000000
## H.npnct26.log 9.890046e-19 <NA> 0.000000
## H.npnct27.log NA <NA> 0.000000
## H.npnct28.log NA <NA> 0.000000
## H.npnct29.log NA <NA> 0.000000
## H.npnct31.log NA <NA> 0.000000
## H.npnct32.log NA <NA> 0.000000
## H.nwrds.unq.log 2.044964e-01 H.nuppr.log 1.019071
## H.T.daili 6.299948e-02 H.T.report 102.903226
## H.T.morn 4.838380e-02 A.npnct30.log 165.205128
## H.T.today 5.831308e-02 H.T.polit 138.239130
## H.T.X2015 6.601141e-02 H.npnct15.log 96.833333
## Popular 1.000000e+00 <NA> 4.976212
## Popular.fctr NA <NA> NA
## PubDate.last1 3.592267e-02 <NA> 1.142857
## PubDate.last10 5.398093e-02 <NA> 1.666667
## PubDate.last100 3.989229e-02 <NA> 25.000000
## PubDate.month.fctr 1.914874e-02 <NA> 1.017514
## PubDate.POSIX 1.568326e-02 <NA> 1.000000
## PubDate.year.fctr NA <NA> 0.000000
## PubDate.zoo 1.568326e-02 <NA> 1.000000
## S.has.http NA <NA> 0.000000
## S.nchrs.log 2.246930e-01 A.nchrs.log 1.328571
## S.npnct02.log 5.547032e-03 <NA> 6531.000000
## S.npnct05.log NA <NA> 0.000000
## S.npnct10.log NA <NA> 0.000000
## S.npnct11.log 5.547032e-03 <NA> 6531.000000
## S.npnct14.log 5.332519e-02 A.npnct14.log 4.672000
## S.npnct17.log 1.587454e-03 <NA> 434.133333
## S.npnct18.log NA <NA> 0.000000
## S.npnct19.log NA <NA> 0.000000
## S.npnct20.log NA <NA> 0.000000
## S.npnct21.log 5.503894e-02 A.npnct21.log 12.862366
## S.npnct23.log 2.760321e-02 A.npnct23.log 6531.000000
## S.npnct24.log NA <NA> 0.000000
## S.npnct25.log 2.760321e-02 <NA> 6531.000000
## S.npnct26.log 9.890046e-19 <NA> 0.000000
## S.npnct27.log NA <NA> 0.000000
## S.npnct28.log NA <NA> 0.000000
## S.npnct29.log NA <NA> 0.000000
## S.npnct31.log NA <NA> 0.000000
## S.npnct32.log NA <NA> 0.000000
## S.nwrds.log 1.359149e-01 A.nwrds.log 2.583333
## S.nwrds.unq.log 2.507969e-01 S.nchrs.log 1.061567
## S.T.articl 5.446201e-02 A.T.articl 85.500000
## S.T.compani 4.764341e-02 A.T.compani 140.227273
## S.T.fashion 8.419711e-02 H.T.X2015 59.809524
## S.T.first 4.617532e-02 A.T.first 203.709677
## S.T.intern 6.953750e-02 A.T.intern 137.347826
## S.T.newyork 5.712853e-02 A.T.newyork 88.724638
## S.T.report 4.746920e-02 A.T.report 78.362500
## S.T.week 8.503373e-02 A.T.week 57.122642
## S.T.will 3.892267e-02 A.T.will 114.750000
## S.T.year 4.729011e-02 A.T.year 167.108108
## UniqueID 1.182492e-02 <NA> 1.000000
## WordCount 2.575265e-01 <NA> 2.315789
## percentUnique zeroVar nzv myNearZV is.cor.y.abs.low
## WordCount.log 24.14268218 FALSE FALSE FALSE FALSE
## myCategory.fctr 0.30618494 FALSE FALSE FALSE FALSE
## H.npnct21.log 0.06123699 FALSE FALSE FALSE FALSE
## A.npnct21.log 0.07654623 FALSE FALSE FALSE FALSE
## S.nuppr.log 0.33680343 FALSE FALSE FALSE FALSE
## H.nchrs.log 1.57685242 FALSE FALSE FALSE FALSE
## A.npnct14.log 0.16840171 FALSE FALSE FALSE FALSE
## H.npnct09.log 0.03061849 FALSE TRUE FALSE FALSE
## PubDate.wkday.fctr 0.10716473 FALSE FALSE FALSE FALSE
## H.nuppr.log 0.29087569 FALSE FALSE FALSE FALSE
## H.T.polit 0.13778322 FALSE TRUE FALSE FALSE
## A.T.newyork 0.42865891 FALSE TRUE FALSE FALSE
## A.T.report 0.38273117 FALSE TRUE FALSE FALSE
## H.npnct12.log 0.07654623 FALSE FALSE FALSE FALSE
## S.ndgts.log 0.26025720 FALSE FALSE FALSE FALSE
## A.nwrds.unq.log 0.55113288 FALSE FALSE FALSE FALSE
## S.T.said 0.36742192 FALSE TRUE FALSE FALSE
## S.T.make 0.47458665 FALSE TRUE FALSE FALSE
## H.npnct17.log 0.06123699 FALSE TRUE FALSE FALSE
## PubDate.last10.log 79.05695040 FALSE FALSE FALSE FALSE
## S.T.state 0.42865891 FALSE TRUE FALSE FALSE
## S.T.one 0.45927740 FALSE TRUE FALSE FALSE
## S.npnct04.log 0.07654623 FALSE TRUE FALSE FALSE
## S.T.share 0.36742192 FALSE TRUE FALSE FALSE
## PubDate.hour.fctr 0.04592774 FALSE FALSE FALSE FALSE
## H.T.report 0.16840171 FALSE TRUE FALSE FALSE
## PubDate.second.fctr 0.06123699 FALSE FALSE FALSE FALSE
## H.T.newyork 0.15309247 FALSE TRUE FALSE FALSE
## S.T.can 0.41334966 FALSE TRUE FALSE FALSE
## H.T.say 0.16840171 FALSE TRUE FALSE FALSE
## A.T.week 0.48989590 FALSE TRUE FALSE FALSE
## H.T.fashion 0.19902021 FALSE TRUE FALSE FALSE
## H.npnct16.log 0.04592774 FALSE FALSE FALSE FALSE
## A.T.will 0.62767912 FALSE TRUE FALSE FALSE
## S.T.show 0.39804042 FALSE TRUE FALSE FALSE
## PubDate.minute.fctr 0.06123699 FALSE FALSE FALSE FALSE
## H.ndgts.log 0.18371096 FALSE FALSE FALSE FALSE
## S.npnct12.log 0.13778322 FALSE FALSE FALSE FALSE
## H.npnct07.log 0.12247397 FALSE FALSE FALSE FALSE
## A.nwrds.log 93.35578690 FALSE FALSE FALSE FALSE
## H.T.art 0.19902021 FALSE TRUE FALSE FALSE
## H.npnct01.log 0.04592774 FALSE TRUE FALSE FALSE
## .rnorm 99.98469075 FALSE FALSE FALSE FALSE
## S.T.time 0.47458665 FALSE TRUE FALSE FALSE
## H.npnct13.log 0.09185548 FALSE FALSE FALSE FALSE
## S.npnct01.log 0.06123699 FALSE TRUE FALSE FALSE
## A.T.intern 0.32149418 FALSE TRUE FALSE FALSE
## A.T.compani 0.50520514 FALSE TRUE FALSE FALSE
## H.T.new 0.19902021 FALSE TRUE FALSE FALSE
## A.nchrs.log 4.39375383 FALSE FALSE FALSE FALSE
## PubDate.last1.log 36.49724434 FALSE FALSE FALSE FALSE
## H.npnct14.log 0.12247397 FALSE TRUE FALSE FALSE
## H.T.first 0.15309247 FALSE TRUE FALSE FALSE
## PubDate.date.fctr 0.07654623 FALSE FALSE FALSE FALSE
## H.T.busi 0.18371096 FALSE TRUE FALSE FALSE
## S.npnct13.log 0.09185548 FALSE FALSE FALSE FALSE
## H.T.day 0.18371096 FALSE TRUE FALSE FALSE
## H.T.china 0.18371096 FALSE TRUE FALSE FALSE
## A.T.year 0.44396816 FALSE TRUE FALSE FALSE
## H.nwrds.log 84.12431108 FALSE FALSE FALSE FALSE
## PubDate.wkend 0.03061849 FALSE FALSE FALSE FALSE
## H.T.big 0.19902021 FALSE TRUE FALSE FALSE
## H.T.news 0.16840171 FALSE TRUE FALSE FALSE
## H.T.week 0.16840171 FALSE TRUE FALSE FALSE
## H.has.ebola 0.03061849 FALSE TRUE FALSE FALSE
## H.T.take 0.15309247 FALSE TRUE FALSE TRUE
## H.T.bank 0.13778322 FALSE TRUE FALSE FALSE
## S.T.take 0.38273117 FALSE TRUE FALSE FALSE
## S.T.day 0.39804042 FALSE TRUE FALSE FALSE
## H.T.make 0.13778322 FALSE TRUE FALSE FALSE
## H.T.time 0.16840171 FALSE TRUE FALSE TRUE
## H.T.pictur 0.10716473 FALSE TRUE FALSE FALSE
## H.T.billion 0.13778322 FALSE TRUE FALSE FALSE
## H.T.X2014 0.13778322 FALSE TRUE FALSE FALSE
## A.T.articl 0.27556644 FALSE TRUE FALSE FALSE
## H.T.obama 0.16840171 FALSE TRUE FALSE FALSE
## S.npnct16.log 0.04592774 FALSE FALSE FALSE FALSE
## PubDate.last100.log 92.19228414 FALSE FALSE FALSE TRUE
## S.npnct15.log 0.04592774 FALSE TRUE FALSE FALSE
## A.T.first 0.42865891 FALSE TRUE FALSE FALSE
## H.T.test 0.13778322 FALSE TRUE FALSE FALSE
## A.npnct17.log 0.04592774 FALSE TRUE FALSE TRUE
## S.npnct06.log 0.03061849 FALSE TRUE FALSE FALSE
## S.T.new 0.48989590 FALSE TRUE FALSE FALSE
## H.npnct30.log 0.03061849 FALSE TRUE FALSE FALSE
## A.T.fashion 0.41334966 FALSE TRUE FALSE FALSE
## H.T.springsumm 0.09185548 FALSE TRUE FALSE FALSE
## H.T.deal 0.13778322 FALSE TRUE FALSE FALSE
## S.npnct30.log 0.04592774 FALSE TRUE FALSE FALSE
## H.npnct02.log 0.03061849 FALSE TRUE FALSE FALSE
## S.npnct22.log 0.03061849 FALSE TRUE FALSE FALSE
## S.npnct07.log 0.04592774 FALSE TRUE FALSE FALSE
## A.T.presid 0.45927740 FALSE TRUE FALSE TRUE
## S.T.presid 0.42865891 FALSE TRUE FALSE TRUE
## S.npnct03.log 0.03061849 FALSE TRUE FALSE FALSE
## S.has.year.colon 0.03061849 FALSE TRUE FALSE FALSE
## H.npnct05.log 0.03061849 FALSE TRUE FALSE FALSE
## S.npnct08.log 0.04592774 FALSE TRUE FALSE TRUE
## S.npnct09.log 0.06123699 FALSE TRUE FALSE TRUE
## A.npnct19.log 0.06123699 FALSE TRUE FALSE FALSE
## A.npnct20.log 0.04592774 FALSE TRUE FALSE FALSE
## A.has.http 0.03061849 FALSE TRUE FALSE FALSE
## A.has.year.colon 0.03061849 FALSE TRUE FALSE FALSE
## A.ndgts.log 0.29087569 FALSE FALSE FALSE FALSE
## A.npnct01.log 0.06123699 FALSE TRUE FALSE FALSE
## A.npnct02.log 0.04592774 FALSE TRUE FALSE FALSE
## A.npnct03.log 0.03061849 FALSE TRUE FALSE FALSE
## A.npnct04.log 0.07654623 FALSE TRUE FALSE FALSE
## A.npnct05.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct06.log 0.03061849 FALSE TRUE FALSE FALSE
## A.npnct07.log 0.04592774 FALSE TRUE FALSE FALSE
## A.npnct08.log 0.04592774 FALSE TRUE FALSE TRUE
## A.npnct09.log 0.06123699 FALSE TRUE FALSE TRUE
## A.npnct10.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct11.log 0.03061849 FALSE TRUE TRUE TRUE
## A.npnct12.log 0.13778322 FALSE FALSE FALSE FALSE
## A.npnct13.log 0.12247397 FALSE FALSE FALSE FALSE
## A.npnct15.log 0.10716473 FALSE TRUE FALSE FALSE
## A.npnct16.log 0.04592774 FALSE FALSE FALSE FALSE
## A.npnct18.log 0.04592774 FALSE TRUE FALSE FALSE
## A.npnct22.log 0.03061849 FALSE TRUE FALSE FALSE
## A.npnct23.log 0.04592774 FALSE TRUE TRUE FALSE
## A.npnct24.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct25.log 0.04592774 FALSE TRUE TRUE FALSE
## A.npnct26.log 0.01530925 TRUE TRUE TRUE TRUE
## A.npnct27.log 0.03061849 FALSE TRUE TRUE TRUE
## A.npnct28.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct29.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct30.log 0.04592774 FALSE TRUE FALSE FALSE
## A.npnct31.log 0.01530925 TRUE TRUE TRUE NA
## A.npnct32.log 0.01530925 TRUE TRUE TRUE NA
## A.nuppr.log 0.33680343 FALSE FALSE FALSE FALSE
## A.T.can 0.48989590 FALSE TRUE FALSE FALSE
## A.T.day 0.42865891 FALSE TRUE FALSE FALSE
## A.T.make 0.48989590 FALSE TRUE FALSE FALSE
## A.T.new 0.50520514 FALSE TRUE FALSE FALSE
## A.T.one 0.50520514 FALSE TRUE FALSE FALSE
## A.T.said 0.39804042 FALSE TRUE FALSE FALSE
## A.T.share 0.36742192 FALSE TRUE FALSE FALSE
## A.T.show 0.41334966 FALSE TRUE FALSE FALSE
## A.T.state 0.42865891 FALSE TRUE FALSE FALSE
## A.T.take 0.42865891 FALSE TRUE FALSE FALSE
## A.T.time 0.47458665 FALSE TRUE FALSE FALSE
## clusterid 0.01530925 TRUE TRUE TRUE NA
## H.has.http 0.01530925 TRUE TRUE TRUE NA
## H.has.year.colon 0.03061849 FALSE TRUE FALSE FALSE
## H.npnct03.log 0.03061849 FALSE TRUE TRUE FALSE
## H.npnct04.log 0.04592774 FALSE TRUE FALSE FALSE
## H.npnct06.log 0.06123699 FALSE TRUE FALSE FALSE
## H.npnct08.log 0.03061849 FALSE TRUE FALSE FALSE
## H.npnct10.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct11.log 0.03061849 FALSE TRUE TRUE TRUE
## H.npnct15.log 0.03061849 FALSE TRUE FALSE FALSE
## H.npnct18.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct19.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct20.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct22.log 0.03061849 FALSE TRUE TRUE TRUE
## H.npnct23.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct24.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct25.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct26.log 0.01530925 TRUE TRUE TRUE TRUE
## H.npnct27.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct28.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct29.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct31.log 0.01530925 TRUE TRUE TRUE NA
## H.npnct32.log 0.01530925 TRUE TRUE TRUE NA
## H.nwrds.unq.log 0.21432945 FALSE FALSE FALSE FALSE
## H.T.daili 0.16840171 FALSE TRUE FALSE FALSE
## H.T.morn 0.07654623 FALSE TRUE FALSE FALSE
## H.T.today 0.13778322 FALSE TRUE FALSE FALSE
## H.T.X2015 0.10716473 FALSE TRUE FALSE FALSE
## Popular 0.03061849 FALSE FALSE FALSE FALSE
## Popular.fctr NA NA NA NA NA
## PubDate.last1 36.49724434 FALSE FALSE FALSE FALSE
## PubDate.last10 79.05695040 FALSE FALSE FALSE FALSE
## PubDate.last100 92.52908757 FALSE FALSE FALSE FALSE
## PubDate.month.fctr 0.04592774 FALSE FALSE FALSE FALSE
## PubDate.POSIX 99.86221678 FALSE FALSE FALSE FALSE
## PubDate.year.fctr 0.01530925 TRUE TRUE TRUE NA
## PubDate.zoo 99.86221678 FALSE FALSE FALSE FALSE
## S.has.http 0.01530925 TRUE TRUE TRUE NA
## S.nchrs.log 3.72014697 FALSE FALSE FALSE FALSE
## S.npnct02.log 0.03061849 FALSE TRUE TRUE TRUE
## S.npnct05.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct10.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct11.log 0.03061849 FALSE TRUE TRUE TRUE
## S.npnct14.log 0.16840171 FALSE FALSE FALSE FALSE
## S.npnct17.log 0.04592774 FALSE TRUE FALSE TRUE
## S.npnct18.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct19.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct20.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct21.log 0.07654623 FALSE FALSE FALSE FALSE
## S.npnct23.log 0.03061849 FALSE TRUE TRUE FALSE
## S.npnct24.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct25.log 0.03061849 FALSE TRUE TRUE FALSE
## S.npnct26.log 0.01530925 TRUE TRUE TRUE TRUE
## S.npnct27.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct28.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct29.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct31.log 0.01530925 TRUE TRUE TRUE NA
## S.npnct32.log 0.01530925 TRUE TRUE TRUE NA
## S.nwrds.log 93.67728108 FALSE FALSE FALSE FALSE
## S.nwrds.unq.log 0.44396816 FALSE FALSE FALSE FALSE
## S.T.articl 0.27556644 FALSE TRUE FALSE FALSE
## S.T.compani 0.44396816 FALSE TRUE FALSE FALSE
## S.T.fashion 0.39804042 FALSE TRUE FALSE FALSE
## S.T.first 0.39804042 FALSE TRUE FALSE FALSE
## S.T.intern 0.30618494 FALSE TRUE FALSE FALSE
## S.T.newyork 0.41334966 FALSE TRUE FALSE FALSE
## S.T.report 0.35211268 FALSE TRUE FALSE FALSE
## S.T.week 0.42865891 FALSE TRUE FALSE FALSE
## S.T.will 0.55113288 FALSE TRUE FALSE FALSE
## S.T.year 0.42865891 FALSE TRUE FALSE FALSE
## UniqueID 100.00000000 FALSE FALSE FALSE FALSE
## WordCount 24.15799143 FALSE FALSE FALSE FALSE
## rsp_var_raw id_var rsp_var importance
## WordCount.log FALSE NA NA 1.000000e+02
## myCategory.fctr FALSE NA NA 7.312143e+01
## H.npnct21.log FALSE NA NA 3.777302e+01
## A.npnct21.log FALSE NA NA 3.475205e+01
## S.nuppr.log FALSE NA NA 3.238555e+01
## H.nchrs.log FALSE NA NA 3.201395e+01
## A.npnct14.log FALSE NA NA 2.764418e+01
## H.npnct09.log FALSE NA NA 2.188969e+01
## PubDate.wkday.fctr FALSE NA NA 2.165613e+01
## H.nuppr.log FALSE NA NA 2.131190e+01
## H.T.polit FALSE NA NA 2.052535e+01
## A.T.newyork FALSE NA NA 1.961261e+01
## A.T.report FALSE NA NA 1.938341e+01
## H.npnct12.log FALSE NA NA 1.796841e+01
## S.ndgts.log FALSE NA NA 1.621363e+01
## A.nwrds.unq.log FALSE NA NA 1.619197e+01
## S.T.said FALSE NA NA 1.520654e+01
## S.T.make FALSE NA NA 1.512129e+01
## H.npnct17.log FALSE NA NA 1.480733e+01
## PubDate.last10.log FALSE NA NA 1.453274e+01
## S.T.state FALSE NA NA 1.414282e+01
## S.T.one FALSE NA NA 1.375899e+01
## S.npnct04.log FALSE NA NA 1.373898e+01
## S.T.share FALSE NA NA 1.314940e+01
## PubDate.hour.fctr FALSE NA NA 1.314384e+01
## H.T.report FALSE NA NA 1.299102e+01
## PubDate.second.fctr FALSE NA NA 1.273156e+01
## H.T.newyork FALSE NA NA 1.269274e+01
## S.T.can FALSE NA NA 1.228026e+01
## H.T.say FALSE NA NA 1.222772e+01
## A.T.week FALSE NA NA 1.168514e+01
## H.T.fashion FALSE NA NA 1.151035e+01
## H.npnct16.log FALSE NA NA 1.136051e+01
## A.T.will FALSE NA NA 1.129252e+01
## S.T.show FALSE NA NA 1.089884e+01
## PubDate.minute.fctr FALSE NA NA 1.041575e+01
## H.ndgts.log FALSE NA NA 1.035074e+01
## S.npnct12.log FALSE NA NA 9.930532e+00
## H.npnct07.log FALSE NA NA 9.854654e+00
## A.nwrds.log FALSE NA NA 9.770236e+00
## H.T.art FALSE NA NA 9.623245e+00
## H.npnct01.log FALSE NA NA 9.509217e+00
## .rnorm FALSE NA NA 9.238260e+00
## S.T.time FALSE NA NA 9.237301e+00
## H.npnct13.log FALSE NA NA 9.109144e+00
## S.npnct01.log FALSE NA NA 8.749827e+00
## A.T.intern FALSE NA NA 8.472657e+00
## A.T.compani FALSE NA NA 8.369682e+00
## H.T.new FALSE NA NA 8.270503e+00
## A.nchrs.log FALSE NA NA 8.248780e+00
## PubDate.last1.log FALSE NA NA 8.146784e+00
## H.npnct14.log FALSE NA NA 8.031911e+00
## H.T.first FALSE NA NA 7.798393e+00
## PubDate.date.fctr FALSE NA NA 7.300560e+00
## H.T.busi FALSE NA NA 7.168757e+00
## S.npnct13.log FALSE NA NA 6.710517e+00
## H.T.day FALSE NA NA 6.378120e+00
## H.T.china FALSE NA NA 6.350450e+00
## A.T.year FALSE NA NA 6.178982e+00
## H.nwrds.log FALSE NA NA 6.165115e+00
## PubDate.wkend FALSE NA NA 6.020520e+00
## H.T.big FALSE NA NA 6.007063e+00
## H.T.news FALSE NA NA 5.524332e+00
## H.T.week FALSE NA NA 5.347019e+00
## H.has.ebola FALSE NA NA 4.853552e+00
## H.T.take FALSE NA NA 4.592147e+00
## H.T.bank FALSE NA NA 4.574012e+00
## S.T.take FALSE NA NA 4.545539e+00
## S.T.day FALSE NA NA 4.501029e+00
## H.T.make FALSE NA NA 4.025030e+00
## H.T.time FALSE NA NA 3.968662e+00
## H.T.pictur FALSE NA NA 3.751942e+00
## H.T.billion FALSE NA NA 3.746828e+00
## H.T.X2014 FALSE NA NA 2.938640e+00
## A.T.articl FALSE NA NA 2.744290e+00
## H.T.obama FALSE NA NA 2.631610e+00
## S.npnct16.log FALSE NA NA 2.586897e+00
## PubDate.last100.log FALSE NA NA 2.562314e+00
## S.npnct15.log FALSE NA NA 1.951937e+00
## A.T.first FALSE NA NA 1.686309e+00
## H.T.test FALSE NA NA 1.393050e+00
## A.npnct17.log FALSE NA NA 1.012142e+00
## S.npnct06.log FALSE NA NA 7.949129e-01
## S.T.new FALSE NA NA 6.136717e-01
## H.npnct30.log FALSE NA NA 5.296067e-01
## A.T.fashion FALSE NA NA 1.689321e-01
## H.T.springsumm FALSE NA NA 8.668136e-02
## H.T.deal FALSE NA NA 6.560495e-02
## S.npnct30.log FALSE NA NA 5.692475e-02
## H.npnct02.log FALSE NA NA 2.826878e-02
## S.npnct22.log FALSE NA NA 2.688801e-02
## S.npnct07.log FALSE NA NA 2.677639e-02
## A.T.presid FALSE NA NA 2.652248e-02
## S.T.presid FALSE NA NA 2.649780e-02
## S.npnct03.log FALSE NA NA 2.562697e-02
## S.has.year.colon FALSE NA NA 2.044323e-02
## H.npnct05.log FALSE NA NA 1.884000e-02
## S.npnct08.log FALSE NA NA 7.744037e-03
## S.npnct09.log FALSE NA NA 6.828413e-03
## A.npnct19.log FALSE NA NA 3.032825e-03
## A.npnct20.log FALSE NA NA 0.000000e+00
## A.has.http FALSE NA NA NA
## A.has.year.colon FALSE NA NA NA
## A.ndgts.log FALSE NA NA NA
## A.npnct01.log FALSE NA NA NA
## A.npnct02.log FALSE NA NA NA
## A.npnct03.log FALSE NA NA NA
## A.npnct04.log FALSE NA NA NA
## A.npnct05.log FALSE NA NA NA
## A.npnct06.log FALSE NA NA NA
## A.npnct07.log FALSE NA NA NA
## A.npnct08.log FALSE NA NA NA
## A.npnct09.log FALSE NA NA NA
## A.npnct10.log FALSE NA NA NA
## A.npnct11.log FALSE NA NA NA
## A.npnct12.log FALSE NA NA NA
## A.npnct13.log FALSE NA NA NA
## A.npnct15.log FALSE NA NA NA
## A.npnct16.log FALSE NA NA NA
## A.npnct18.log FALSE NA NA NA
## A.npnct22.log FALSE NA NA NA
## A.npnct23.log FALSE NA NA NA
## A.npnct24.log FALSE NA NA NA
## A.npnct25.log FALSE NA NA NA
## A.npnct26.log FALSE NA NA NA
## A.npnct27.log FALSE NA NA NA
## A.npnct28.log FALSE NA NA NA
## A.npnct29.log FALSE NA NA NA
## A.npnct30.log FALSE NA NA NA
## A.npnct31.log FALSE NA NA NA
## A.npnct32.log FALSE NA NA NA
## A.nuppr.log FALSE NA NA NA
## A.T.can FALSE NA NA NA
## A.T.day FALSE NA NA NA
## A.T.make FALSE NA NA NA
## A.T.new FALSE NA NA NA
## A.T.one FALSE NA NA NA
## A.T.said FALSE NA NA NA
## A.T.share FALSE NA NA NA
## A.T.show FALSE NA NA NA
## A.T.state FALSE NA NA NA
## A.T.take FALSE NA NA NA
## A.T.time FALSE NA NA NA
## clusterid FALSE NA NA NA
## H.has.http FALSE NA NA NA
## H.has.year.colon FALSE NA NA NA
## H.npnct03.log FALSE NA NA NA
## H.npnct04.log FALSE NA NA NA
## H.npnct06.log FALSE NA NA NA
## H.npnct08.log FALSE NA NA NA
## H.npnct10.log FALSE NA NA NA
## H.npnct11.log FALSE NA NA NA
## H.npnct15.log FALSE NA NA NA
## H.npnct18.log FALSE NA NA NA
## H.npnct19.log FALSE NA NA NA
## H.npnct20.log FALSE NA NA NA
## H.npnct22.log FALSE NA NA NA
## H.npnct23.log FALSE NA NA NA
## H.npnct24.log FALSE NA NA NA
## H.npnct25.log FALSE NA NA NA
## H.npnct26.log FALSE NA NA NA
## H.npnct27.log FALSE NA NA NA
## H.npnct28.log FALSE NA NA NA
## H.npnct29.log FALSE NA NA NA
## H.npnct31.log FALSE NA NA NA
## H.npnct32.log FALSE NA NA NA
## H.nwrds.unq.log FALSE NA NA NA
## H.T.daili FALSE NA NA NA
## H.T.morn FALSE NA NA NA
## H.T.today FALSE NA NA NA
## H.T.X2015 FALSE NA NA NA
## Popular TRUE NA NA NA
## Popular.fctr NA NA TRUE NA
## PubDate.last1 FALSE NA NA NA
## PubDate.last10 FALSE NA NA NA
## PubDate.last100 FALSE NA NA NA
## PubDate.month.fctr FALSE NA NA NA
## PubDate.POSIX FALSE NA NA NA
## PubDate.year.fctr FALSE NA NA NA
## PubDate.zoo FALSE NA NA NA
## S.has.http FALSE NA NA NA
## S.nchrs.log FALSE NA NA NA
## S.npnct02.log FALSE NA NA NA
## S.npnct05.log FALSE NA NA NA
## S.npnct10.log FALSE NA NA NA
## S.npnct11.log FALSE NA NA NA
## S.npnct14.log FALSE NA NA NA
## S.npnct17.log FALSE NA NA NA
## S.npnct18.log FALSE NA NA NA
## S.npnct19.log FALSE NA NA NA
## S.npnct20.log FALSE NA NA NA
## S.npnct21.log FALSE NA NA NA
## S.npnct23.log FALSE NA NA NA
## S.npnct24.log FALSE NA NA NA
## S.npnct25.log FALSE NA NA NA
## S.npnct26.log FALSE NA NA NA
## S.npnct27.log FALSE NA NA NA
## S.npnct28.log FALSE NA NA NA
## S.npnct29.log FALSE NA NA NA
## S.npnct31.log FALSE NA NA NA
## S.npnct32.log FALSE NA NA NA
## S.nwrds.log FALSE NA NA NA
## S.nwrds.unq.log FALSE NA NA NA
## S.T.articl FALSE NA NA NA
## S.T.compani FALSE NA NA NA
## S.T.fashion FALSE NA NA NA
## S.T.first FALSE NA NA NA
## S.T.intern FALSE NA NA NA
## S.T.newyork FALSE NA NA NA
## S.T.report FALSE NA NA NA
## S.T.week FALSE NA NA NA
## S.T.will FALSE NA NA NA
## S.T.year FALSE NA NA NA
## UniqueID FALSE TRUE NA NA
## WordCount FALSE NA NA NA
## Low.cor.X.glm.importance
## WordCount.log 1.000000e+02
## myCategory.fctr 7.312143e+01
## H.npnct21.log 3.777302e+01
## A.npnct21.log 3.475205e+01
## S.nuppr.log 3.238555e+01
## H.nchrs.log 3.201395e+01
## A.npnct14.log 2.764418e+01
## H.npnct09.log 2.188969e+01
## PubDate.wkday.fctr 2.165613e+01
## H.nuppr.log 2.131190e+01
## H.T.polit 2.052535e+01
## A.T.newyork 1.961261e+01
## A.T.report 1.938341e+01
## H.npnct12.log 1.796841e+01
## S.ndgts.log 1.621363e+01
## A.nwrds.unq.log 1.619197e+01
## S.T.said 1.520654e+01
## S.T.make 1.512129e+01
## H.npnct17.log 1.480733e+01
## PubDate.last10.log 1.453274e+01
## S.T.state 1.414282e+01
## S.T.one 1.375899e+01
## S.npnct04.log 1.373898e+01
## S.T.share 1.314940e+01
## PubDate.hour.fctr 1.314384e+01
## H.T.report 1.299102e+01
## PubDate.second.fctr 1.273156e+01
## H.T.newyork 1.269274e+01
## S.T.can 1.228026e+01
## H.T.say 1.222772e+01
## A.T.week 1.168514e+01
## H.T.fashion 1.151035e+01
## H.npnct16.log 1.136051e+01
## A.T.will 1.129252e+01
## S.T.show 1.089884e+01
## PubDate.minute.fctr 1.041575e+01
## H.ndgts.log 1.035074e+01
## S.npnct12.log 9.930532e+00
## H.npnct07.log 9.854654e+00
## A.nwrds.log 9.770236e+00
## H.T.art 9.623245e+00
## H.npnct01.log 9.509217e+00
## .rnorm 9.238260e+00
## S.T.time 9.237301e+00
## H.npnct13.log 9.109144e+00
## S.npnct01.log 8.749827e+00
## A.T.intern 8.472657e+00
## A.T.compani 8.369682e+00
## H.T.new 8.270503e+00
## A.nchrs.log 8.248780e+00
## PubDate.last1.log 8.146784e+00
## H.npnct14.log 8.031911e+00
## H.T.first 7.798393e+00
## PubDate.date.fctr 7.300560e+00
## H.T.busi 7.168757e+00
## S.npnct13.log 6.710517e+00
## H.T.day 6.378120e+00
## H.T.china 6.350450e+00
## A.T.year 6.178982e+00
## H.nwrds.log 6.165115e+00
## PubDate.wkend 6.020520e+00
## H.T.big 6.007063e+00
## H.T.news 5.524332e+00
## H.T.week 5.347019e+00
## H.has.ebola 4.853552e+00
## H.T.take 4.592147e+00
## H.T.bank 4.574012e+00
## S.T.take 4.545539e+00
## S.T.day 4.501029e+00
## H.T.make 4.025030e+00
## H.T.time 3.968662e+00
## H.T.pictur 3.751942e+00
## H.T.billion 3.746828e+00
## H.T.X2014 2.938640e+00
## A.T.articl 2.744290e+00
## H.T.obama 2.631610e+00
## S.npnct16.log 2.586897e+00
## PubDate.last100.log 2.562314e+00
## S.npnct15.log 1.951937e+00
## A.T.first 1.686309e+00
## H.T.test 1.393050e+00
## A.npnct17.log 1.012142e+00
## S.npnct06.log 7.949129e-01
## S.T.new 6.136717e-01
## H.npnct30.log 5.296067e-01
## A.T.fashion 1.689321e-01
## H.T.springsumm 8.668136e-02
## H.T.deal 6.560495e-02
## S.npnct30.log 5.692475e-02
## H.npnct02.log 2.826878e-02
## S.npnct22.log 2.688801e-02
## S.npnct07.log 2.677639e-02
## A.T.presid 2.652248e-02
## S.T.presid 2.649780e-02
## S.npnct03.log 2.562697e-02
## S.has.year.colon 2.044323e-02
## H.npnct05.log 1.884000e-02
## S.npnct08.log 7.744037e-03
## S.npnct09.log 6.828413e-03
## A.npnct19.log 3.032825e-03
## A.npnct20.log 0.000000e+00
## A.has.http NA
## A.has.year.colon NA
## A.ndgts.log NA
## A.npnct01.log NA
## A.npnct02.log NA
## A.npnct03.log NA
## A.npnct04.log NA
## A.npnct05.log NA
## A.npnct06.log NA
## A.npnct07.log NA
## A.npnct08.log NA
## A.npnct09.log NA
## A.npnct10.log NA
## A.npnct11.log NA
## A.npnct12.log NA
## A.npnct13.log NA
## A.npnct15.log NA
## A.npnct16.log NA
## A.npnct18.log NA
## A.npnct22.log NA
## A.npnct23.log NA
## A.npnct24.log NA
## A.npnct25.log NA
## A.npnct26.log NA
## A.npnct27.log NA
## A.npnct28.log NA
## A.npnct29.log NA
## A.npnct30.log NA
## A.npnct31.log NA
## A.npnct32.log NA
## A.nuppr.log NA
## A.T.can NA
## A.T.day NA
## A.T.make NA
## A.T.new NA
## A.T.one NA
## A.T.said NA
## A.T.share NA
## A.T.show NA
## A.T.state NA
## A.T.take NA
## A.T.time NA
## clusterid NA
## H.has.http NA
## H.has.year.colon NA
## H.npnct03.log NA
## H.npnct04.log NA
## H.npnct06.log NA
## H.npnct08.log NA
## H.npnct10.log NA
## H.npnct11.log NA
## H.npnct15.log NA
## H.npnct18.log NA
## H.npnct19.log NA
## H.npnct20.log NA
## H.npnct22.log NA
## H.npnct23.log NA
## H.npnct24.log NA
## H.npnct25.log NA
## H.npnct26.log NA
## H.npnct27.log NA
## H.npnct28.log NA
## H.npnct29.log NA
## H.npnct31.log NA
## H.npnct32.log NA
## H.nwrds.unq.log NA
## H.T.daili NA
## H.T.morn NA
## H.T.today NA
## H.T.X2015 NA
## Popular NA
## Popular.fctr NA
## PubDate.last1 NA
## PubDate.last10 NA
## PubDate.last100 NA
## PubDate.month.fctr NA
## PubDate.POSIX NA
## PubDate.year.fctr NA
## PubDate.zoo NA
## S.has.http NA
## S.nchrs.log NA
## S.npnct02.log NA
## S.npnct05.log NA
## S.npnct10.log NA
## S.npnct11.log NA
## S.npnct14.log NA
## S.npnct17.log NA
## S.npnct18.log NA
## S.npnct19.log NA
## S.npnct20.log NA
## S.npnct21.log NA
## S.npnct23.log NA
## S.npnct24.log NA
## S.npnct25.log NA
## S.npnct26.log NA
## S.npnct27.log NA
## S.npnct28.log NA
## S.npnct29.log NA
## S.npnct31.log NA
## S.npnct32.log NA
## S.nwrds.log NA
## S.nwrds.unq.log NA
## S.T.articl NA
## S.T.compani NA
## S.T.fashion NA
## S.T.first NA
## S.T.intern NA
## S.T.newyork NA
## S.T.report NA
## S.T.week NA
## S.T.will NA
## S.T.year NA
## UniqueID NA
## WordCount NA# Used again in fit.data.training & predict.data.new chunks
glb_analytics_diag_plots <- function(obs_df, mdl_id, prob_threshold=NULL) {
if (length(vars <- subset(glb_feats_df, importance > 0)$id) > 5) {
warning("Limiting important feature scatter plots to 5 out of ", length(vars))
vars <- vars[1:5]
}
require(reshape2)
rsp_var_out <- paste0(glb_rsp_var_out, mdl_id)
for (var in vars) {
plot_df <- melt(obs_df, id.vars=var,
measure.vars=c(glb_rsp_var, rsp_var_out))
# if (var == "<feat_name>") print(myplot_scatter(plot_df, var, "value",
# facet_colcol_name="variable") +
# geom_vline(xintercept=<divider_val>, linetype="dotted")) else
print(myplot_scatter(plot_df, var, "value", colorcol_name="variable",
facet_colcol_name="variable", jitter=TRUE) +
guides(color=FALSE))
}
if (glb_is_regression) {
# plot_vars_df <- subset(glb_feats_df, importance >
# glb_feats_df[glb_feats_df$id == ".rnorm", "importance"])
plot_vars_df <- orderBy(~ -importance, glb_feats_df)
if (nrow(plot_vars_df) == 0)
warning("No important features in glb_fin_mdl") else
print(myplot_prediction_regression(df=obs_df,
feat_x=ifelse(nrow(plot_vars_df) > 1, plot_vars_df$id[2],
".rownames"),
feat_y=plot_vars_df$id[1],
rsp_var=glb_rsp_var, rsp_var_out=rsp_var_out,
id_vars=glb_id_vars)
# + facet_wrap(reformulate(plot_vars_df$id[2])) # if [1 or 2] is a factor
# + geom_point(aes_string(color="<col_name>.fctr")) # to color the plot
)
}
if (glb_is_classification) {
if (nrow(plot_vars_df <- subset(glb_feats_df, importance > 0)) == 0)
warning("No features in selected model are statistically important")
else print(myplot_prediction_classification(df=obs_df,
feat_x=ifelse(nrow(plot_vars_df) > 1, plot_vars_df$id[2],
".rownames"),
feat_y=plot_vars_df$id[1],
rsp_var=glb_rsp_var,
rsp_var_out=rsp_var_out,
id_vars=glb_id_vars,
prob_threshold=prob_threshold)
# + geom_hline(yintercept=<divider_val>, linetype = "dotted")
)
}
}
glb_analytics_diag_plots(obs_df=glb_OOBent_df, mdl_id=glb_sel_mdl_id,
prob_threshold=ifelse(glb_is_classification && glb_is_binomial,
glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"], NULL))## Warning in glb_analytics_diag_plots(obs_df = glb_OOBent_df, mdl_id =
## glb_sel_mdl_id, : Limiting important feature scatter plots to 5 out of 100
## [1] "Min/Max Boundaries: "
## UniqueID Popular.fctr Popular.fctr.predict.Low.cor.X.glm.prob
## 6018 6018 N 0.0002963389
## 6370 6370 Y 0.4783260065
## Popular.fctr.predict.Low.cor.X.glm
## 6018 N
## 6370 Y
## Popular.fctr.predict.Low.cor.X.glm.accurate
## 6018 TRUE
## 6370 TRUE
## Popular.fctr.predict.Low.cor.X.glm.error .label
## 6018 0 6018
## 6370 0 6370
## [1] "Inaccurate: "
## UniqueID Popular.fctr Popular.fctr.predict.Low.cor.X.glm.prob
## 3743 3743 Y 2.220446e-16
## 5423 5423 Y 2.220446e-16
## 5486 5486 Y 1.056241e-10
## 5573 5573 Y 1.093977e-10
## 2026 2026 Y 3.529691e-10
## 6387 6387 Y 5.867900e-10
## Popular.fctr.predict.Low.cor.X.glm
## 3743 N
## 5423 N
## 5486 N
## 5573 N
## 2026 N
## 6387 N
## Popular.fctr.predict.Low.cor.X.glm.accurate
## 3743 FALSE
## 5423 FALSE
## 5486 FALSE
## 5573 FALSE
## 2026 FALSE
## 6387 FALSE
## Popular.fctr.predict.Low.cor.X.glm.error
## 3743 -0.3
## 5423 -0.3
## 5486 -0.3
## 5573 -0.3
## 2026 -0.3
## 6387 -0.3
## UniqueID Popular.fctr Popular.fctr.predict.Low.cor.X.glm.prob
## 1096 1096 Y 0.1271185
## 3670 3670 N 0.4067999
## 1677 1677 N 0.6875558
## 4597 4597 N 0.7137185
## 3875 3875 N 0.7741152
## 5244 5244 N 0.9002041
## Popular.fctr.predict.Low.cor.X.glm
## 1096 N
## 3670 Y
## 1677 Y
## 4597 Y
## 3875 Y
## 5244 Y
## Popular.fctr.predict.Low.cor.X.glm.accurate
## 1096 FALSE
## 3670 FALSE
## 1677 FALSE
## 4597 FALSE
## 3875 FALSE
## 5244 FALSE
## Popular.fctr.predict.Low.cor.X.glm.error
## 1096 -0.1728815
## 3670 0.1067999
## 1677 0.3875558
## 4597 0.4137185
## 3875 0.4741152
## 5244 0.6002041
## UniqueID Popular.fctr Popular.fctr.predict.Low.cor.X.glm.prob
## 4975 4975 N 0.9387725
## 3258 3258 N 0.9431130
## 4771 4771 N 0.9445898
## 1667 1667 N 0.9579650
## 4882 4882 N 0.9817055
## 770 770 N 0.9849727
## Popular.fctr.predict.Low.cor.X.glm
## 4975 Y
## 3258 Y
## 4771 Y
## 1667 Y
## 4882 Y
## 770 Y
## Popular.fctr.predict.Low.cor.X.glm.accurate
## 4975 FALSE
## 3258 FALSE
## 4771 FALSE
## 1667 FALSE
## 4882 FALSE
## 770 FALSE
## Popular.fctr.predict.Low.cor.X.glm.error
## 4975 0.6387725
## 3258 0.6431130
## 4771 0.6445898
## 1667 0.6579650
## 4882 0.6817055
## 770 0.6849727
# gather predictions from models better than MFO.*
#mdl_id <- "Conditional.X.rf"
#mdl_id <- "Conditional.X.cp.0.rpart"
#mdl_id <- "Conditional.X.rpart"
# glb_OOBent_df <- glb_get_predictions(df=glb_OOBent_df, mdl_id,
# glb_rsp_var_out)
# print(t(confusionMatrix(glb_OOBent_df[, paste0(glb_rsp_var_out, mdl_id)],
# glb_OOBent_df[, glb_rsp_var])$table))
FN_OOB_ids <- c(4721, 4020, 693, 92)
print(glb_OOBent_df[glb_OOBent_df$UniqueID %in% FN_OOB_ids,
grep(glb_rsp_var, names(glb_OOBent_df), value=TRUE)])## [1] Popular.fctr
## [2] Popular.fctr.predict.Low.cor.X.glm.prob
## [3] Popular.fctr.predict.Low.cor.X.glm
## [4] Popular.fctr.predict.Low.cor.X.glm.accurate
## <0 rows> (or 0-length row.names)print(glb_OOBent_df[glb_OOBent_df$UniqueID %in% FN_OOB_ids,
glb_feats_df$id[1:5]])## [1] WordCount.log myCategory.fctr H.npnct21.log A.npnct21.log
## [5] S.nuppr.log
## <0 rows> (or 0-length row.names)print(glb_OOBent_df[glb_OOBent_df$UniqueID %in% FN_OOB_ids,
glb_txt_vars])## [1] Headline Snippet Abstract
## <0 rows> (or 0-length row.names)write.csv(glb_OOBent_df[, c("UniqueID",
grep(glb_rsp_var, names(glb_OOBent_df), fixed=TRUE, value=TRUE))],
paste0(gsub(".", "_", paste0(glb_out_pfx, glb_sel_mdl_id), fixed=TRUE),
"_OOBent.csv"), row.names=FALSE)
# print(glb_entity_df[glb_entity_df$UniqueID %in% FN_OOB_ids,
# glb_txt_vars])
# dsp_tbl(Headline.contains="[Ee]bola")
# sum(sel_obs(Headline.contains="[Ee]bola"))
# ftable(xtabs(Popular ~ NewsDesk.fctr, data=glb_entity_df[sel_obs(Headline.contains="[Ee]bola") ,]))
# xtabs(NewsDesk ~ Popular, #Popular ~ NewsDesk.fctr,
# data=glb_entity_df[sel_obs(Headline.contains="[Ee]bola") ,],
# exclude=NULL)
# print(mycreate_xtab_df(df=glb_entity_df[sel_obs(Headline.contains="[Ee]bola") ,], c("Popular", "NewsDesk", "SectionName", "SubsectionName")))
# print(mycreate_tbl_df(df=glb_entity_df[sel_obs(Headline.contains="[Ee]bola") ,], c("Popular", "NewsDesk", "SectionName", "SubsectionName")))
# print(mycreate_tbl_df(df=glb_entity_df[sel_obs(Headline.contains="[Ee]bola") ,], c("Popular")))
# print(mycreate_tbl_df(df=glb_entity_df[sel_obs(Headline.contains="[Ee]bola") ,],
# tbl_col_names=c("Popular", "NewsDesk")))
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 12 fit.models 7 2 312.670 328.058 15.388
## 13 fit.models 7 3 328.059 NA NAsav_entity_df <- glb_entity_df
print(setdiff(names(glb_trnent_df), names(glb_entity_df)))## [1] "PubDate.year.fctr" "H.has.http" "H.npnct03.log"
## [4] "H.npnct10.log" "H.npnct11.log" "H.npnct18.log"
## [7] "H.npnct19.log" "H.npnct20.log" "H.npnct22.log"
## [10] "H.npnct23.log" "H.npnct24.log" "H.npnct25.log"
## [13] "H.npnct26.log" "H.npnct27.log" "H.npnct28.log"
## [16] "H.npnct29.log" "H.npnct31.log" "H.npnct32.log"
## [19] "S.has.http" "S.npnct02.log" "S.npnct05.log"
## [22] "S.npnct10.log" "S.npnct11.log" "S.npnct18.log"
## [25] "S.npnct19.log" "S.npnct20.log" "S.npnct23.log"
## [28] "S.npnct24.log" "S.npnct25.log" "S.npnct26.log"
## [31] "S.npnct27.log" "S.npnct28.log" "S.npnct29.log"
## [34] "S.npnct31.log" "S.npnct32.log" "A.npnct05.log"
## [37] "A.npnct10.log" "A.npnct11.log" "A.npnct23.log"
## [40] "A.npnct24.log" "A.npnct25.log" "A.npnct26.log"
## [43] "A.npnct27.log" "A.npnct28.log" "A.npnct29.log"
## [46] "A.npnct31.log" "A.npnct32.log" "clusterid"print(setdiff(names(glb_fitent_df), names(glb_entity_df)))## [1] "PubDate.year.fctr" "H.has.http" "H.npnct03.log"
## [4] "H.npnct10.log" "H.npnct11.log" "H.npnct18.log"
## [7] "H.npnct19.log" "H.npnct20.log" "H.npnct22.log"
## [10] "H.npnct23.log" "H.npnct24.log" "H.npnct25.log"
## [13] "H.npnct26.log" "H.npnct27.log" "H.npnct28.log"
## [16] "H.npnct29.log" "H.npnct31.log" "H.npnct32.log"
## [19] "S.has.http" "S.npnct02.log" "S.npnct05.log"
## [22] "S.npnct10.log" "S.npnct11.log" "S.npnct18.log"
## [25] "S.npnct19.log" "S.npnct20.log" "S.npnct23.log"
## [28] "S.npnct24.log" "S.npnct25.log" "S.npnct26.log"
## [31] "S.npnct27.log" "S.npnct28.log" "S.npnct29.log"
## [34] "S.npnct31.log" "S.npnct32.log" "A.npnct05.log"
## [37] "A.npnct10.log" "A.npnct11.log" "A.npnct23.log"
## [40] "A.npnct24.log" "A.npnct25.log" "A.npnct26.log"
## [43] "A.npnct27.log" "A.npnct28.log" "A.npnct29.log"
## [46] "A.npnct31.log" "A.npnct32.log" "clusterid"print(setdiff(names(glb_OOBent_df), names(glb_entity_df)))## [1] "PubDate.year.fctr"
## [2] "H.has.http"
## [3] "H.npnct03.log"
## [4] "H.npnct10.log"
## [5] "H.npnct11.log"
## [6] "H.npnct18.log"
## [7] "H.npnct19.log"
## [8] "H.npnct20.log"
## [9] "H.npnct22.log"
## [10] "H.npnct23.log"
## [11] "H.npnct24.log"
## [12] "H.npnct25.log"
## [13] "H.npnct26.log"
## [14] "H.npnct27.log"
## [15] "H.npnct28.log"
## [16] "H.npnct29.log"
## [17] "H.npnct31.log"
## [18] "H.npnct32.log"
## [19] "S.has.http"
## [20] "S.npnct02.log"
## [21] "S.npnct05.log"
## [22] "S.npnct10.log"
## [23] "S.npnct11.log"
## [24] "S.npnct18.log"
## [25] "S.npnct19.log"
## [26] "S.npnct20.log"
## [27] "S.npnct23.log"
## [28] "S.npnct24.log"
## [29] "S.npnct25.log"
## [30] "S.npnct26.log"
## [31] "S.npnct27.log"
## [32] "S.npnct28.log"
## [33] "S.npnct29.log"
## [34] "S.npnct31.log"
## [35] "S.npnct32.log"
## [36] "A.npnct05.log"
## [37] "A.npnct10.log"
## [38] "A.npnct11.log"
## [39] "A.npnct23.log"
## [40] "A.npnct24.log"
## [41] "A.npnct25.log"
## [42] "A.npnct26.log"
## [43] "A.npnct27.log"
## [44] "A.npnct28.log"
## [45] "A.npnct29.log"
## [46] "A.npnct31.log"
## [47] "A.npnct32.log"
## [48] "clusterid"
## [49] "Popular.fctr.predict.Low.cor.X.glm.prob"
## [50] "Popular.fctr.predict.Low.cor.X.glm"
## [51] "Popular.fctr.predict.Low.cor.X.glm.accurate"for (col in setdiff(names(glb_OOBent_df), names(glb_entity_df)))
# Merge or cbind ?
glb_entity_df[glb_entity_df$.lcn == "OOB", col] <- glb_OOBent_df[, col]
print(setdiff(names(glb_newent_df), names(glb_entity_df)))## character(0)if (glb_save_envir)
save(glb_feats_df,
glb_entity_df, #glb_trnent_df, glb_fitent_df, glb_OOBent_df, glb_newent_df,
glb_models_df, dsp_models_df, glb_models_lst, glb_sel_mdl, glb_sel_mdl_id,
glb_model_type,
file=paste0(glb_out_pfx, "selmdl_dsk.RData"))
#load(paste0(glb_out_pfx, "selmdl_dsk.RData"))
replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"model.selected")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.data.training", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 13 fit.models 7 3 328.059 335.358 7.299
## 14 fit.data.training 8 0 335.359 NA NAStep 8.0: fit data training
#load(paste0(glb_inp_pfx, "dsk.RData"))
# To create specific models
# glb_fin_mdl_id <- NULL; glb_fin_mdl <- NULL;
# glb_sel_mdl_id <- "Conditional.X.cp.0.rpart";
# glb_sel_mdl <- glb_models_lst[[glb_sel_mdl_id]]; print(glb_sel_mdl)
if (!is.null(glb_fin_mdl_id) && (glb_fin_mdl_id %in% names(glb_models_lst))) {
warning("Final model same as user selected model")
glb_fin_mdl <- glb_sel_mdl
} else {
print(mdl_feats_df <- myextract_mdl_feats(sel_mdl=glb_sel_mdl,
entity_df=glb_fitent_df))
if ((model_method <- glb_sel_mdl$method) == "custom")
# get actual method from the model_id
model_method <- tail(unlist(strsplit(glb_sel_mdl_id, "[.]")), 1)
tune_finmdl_df <- NULL
if (nrow(glb_sel_mdl$bestTune) > 0) {
for (param in names(glb_sel_mdl$bestTune)) {
#print(sprintf("param: %s", param))
if (glb_sel_mdl$bestTune[1, param] != "none")
tune_finmdl_df <- rbind(tune_finmdl_df,
data.frame(parameter=param,
min=glb_sel_mdl$bestTune[1, param],
max=glb_sel_mdl$bestTune[1, param],
by=1)) # by val does not matter
}
}
# Sync with parameters in mydsutils.R
ret_lst <- myfit_mdl(model_id="Final", model_method=model_method,
indep_vars_vctr=mdl_feats_df$id, model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=NULL,
n_cv_folds=glb_n_cv_folds, tune_models_df=tune_finmdl_df,
# Automate from here
# Issues if glb_sel_mdl$method == "rf" b/c trainControl is "oob"; not "cv"
model_loss_mtrx=glb_model_metric_terms,
model_summaryFunction=glb_sel_mdl$control$summaryFunction,
model_metric=glb_sel_mdl$metric,
model_metric_maximize=glb_sel_mdl$maximize)
glb_fin_mdl <- glb_models_lst[[length(glb_models_lst)]]
glb_fin_mdl_id <- glb_models_df[length(glb_models_lst), "model_id"]
}## id importance
## WordCount.log WordCount.log 1.000000e+02
## myCategory.fctr myCategory.fctr 7.312143e+01
## H.npnct21.log H.npnct21.log 3.777302e+01
## A.npnct21.log A.npnct21.log 3.475205e+01
## S.nuppr.log S.nuppr.log 3.238555e+01
## H.nchrs.log H.nchrs.log 3.201395e+01
## A.npnct14.log A.npnct14.log 2.764418e+01
## H.npnct09.log H.npnct09.log 2.188969e+01
## PubDate.wkday.fctr PubDate.wkday.fctr 2.165613e+01
## H.nuppr.log H.nuppr.log 2.131190e+01
## H.T.polit H.T.polit 2.052535e+01
## A.T.newyork A.T.newyork 1.961261e+01
## A.T.report A.T.report 1.938341e+01
## H.npnct12.log H.npnct12.log 1.796841e+01
## S.ndgts.log S.ndgts.log 1.621363e+01
## A.nwrds.unq.log A.nwrds.unq.log 1.619197e+01
## S.T.said S.T.said 1.520654e+01
## S.T.make S.T.make 1.512129e+01
## H.npnct17.log H.npnct17.log 1.480733e+01
## PubDate.last10.log PubDate.last10.log 1.453274e+01
## S.T.state S.T.state 1.414282e+01
## S.T.one S.T.one 1.375899e+01
## S.npnct04.log S.npnct04.log 1.373898e+01
## S.T.share S.T.share 1.314940e+01
## PubDate.hour.fctr PubDate.hour.fctr 1.314384e+01
## H.T.report H.T.report 1.299102e+01
## PubDate.second.fctr PubDate.second.fctr 1.273156e+01
## H.T.newyork H.T.newyork 1.269274e+01
## S.T.can S.T.can 1.228026e+01
## H.T.say H.T.say 1.222772e+01
## A.T.week A.T.week 1.168514e+01
## H.T.fashion H.T.fashion 1.151035e+01
## H.npnct16.log H.npnct16.log 1.136051e+01
## A.T.will A.T.will 1.129252e+01
## S.T.show S.T.show 1.089884e+01
## PubDate.minute.fctr PubDate.minute.fctr 1.041575e+01
## H.ndgts.log H.ndgts.log 1.035074e+01
## S.npnct12.log S.npnct12.log 9.930532e+00
## H.npnct07.log H.npnct07.log 9.854654e+00
## A.nwrds.log A.nwrds.log 9.770236e+00
## H.T.art H.T.art 9.623245e+00
## H.npnct01.log H.npnct01.log 9.509217e+00
## .rnorm .rnorm 9.238260e+00
## S.T.time S.T.time 9.237301e+00
## H.npnct13.log H.npnct13.log 9.109144e+00
## S.npnct01.log S.npnct01.log 8.749827e+00
## A.T.intern A.T.intern 8.472657e+00
## A.T.compani A.T.compani 8.369682e+00
## H.T.new H.T.new 8.270503e+00
## A.nchrs.log A.nchrs.log 8.248780e+00
## PubDate.last1.log PubDate.last1.log 8.146784e+00
## H.npnct14.log H.npnct14.log 8.031911e+00
## H.T.first H.T.first 7.798393e+00
## PubDate.date.fctr PubDate.date.fctr 7.300560e+00
## H.T.busi H.T.busi 7.168757e+00
## S.npnct13.log S.npnct13.log 6.710517e+00
## H.T.day H.T.day 6.378120e+00
## H.T.china H.T.china 6.350450e+00
## A.T.year A.T.year 6.178982e+00
## H.nwrds.log H.nwrds.log 6.165115e+00
## PubDate.wkend PubDate.wkend 6.020520e+00
## H.T.big H.T.big 6.007063e+00
## H.T.news H.T.news 5.524332e+00
## H.T.week H.T.week 5.347019e+00
## H.has.ebola H.has.ebola 4.853552e+00
## H.T.take H.T.take 4.592147e+00
## H.T.bank H.T.bank 4.574012e+00
## S.T.take S.T.take 4.545539e+00
## S.T.day S.T.day 4.501029e+00
## H.T.make H.T.make 4.025030e+00
## H.T.time H.T.time 3.968662e+00
## H.T.pictur H.T.pictur 3.751942e+00
## H.T.billion H.T.billion 3.746828e+00
## H.T.X2014 H.T.X2014 2.938640e+00
## A.T.articl A.T.articl 2.744290e+00
## H.T.obama H.T.obama 2.631610e+00
## S.npnct16.log S.npnct16.log 2.586897e+00
## PubDate.last100.log PubDate.last100.log 2.562314e+00
## S.npnct15.log S.npnct15.log 1.951937e+00
## A.T.first A.T.first 1.686309e+00
## H.T.test H.T.test 1.393050e+00
## A.npnct17.log A.npnct17.log 1.012142e+00
## S.npnct06.log S.npnct06.log 7.949129e-01
## S.T.new S.T.new 6.136717e-01
## H.npnct30.log H.npnct30.log 5.296067e-01
## A.T.fashion A.T.fashion 1.689321e-01
## H.T.springsumm H.T.springsumm 8.668136e-02
## H.T.deal H.T.deal 6.560495e-02
## S.npnct30.log S.npnct30.log 5.692475e-02
## H.npnct02.log H.npnct02.log 2.826878e-02
## S.npnct22.log S.npnct22.log 2.688801e-02
## S.npnct07.log S.npnct07.log 2.677639e-02
## A.T.presid A.T.presid 2.652248e-02
## S.T.presid S.T.presid 2.649780e-02
## S.npnct03.log S.npnct03.log 2.562697e-02
## S.has.year.colon S.has.year.colon 2.044323e-02
## H.npnct05.log H.npnct05.log 1.884000e-02
## S.npnct08.log S.npnct08.log 7.744037e-03
## S.npnct09.log S.npnct09.log 6.828413e-03
## A.npnct19.log A.npnct19.log 3.032825e-03
## A.npnct20.log A.npnct20.log 0.000000e+00
## [1] "fitting model: Final.glm"
## [1] " indep_vars: WordCount.log, myCategory.fctr, H.npnct21.log, A.npnct21.log, S.nuppr.log, H.nchrs.log, A.npnct14.log, H.npnct09.log, PubDate.wkday.fctr, H.nuppr.log, H.T.polit, A.T.newyork, A.T.report, H.npnct12.log, S.ndgts.log, A.nwrds.unq.log, S.T.said, S.T.make, H.npnct17.log, PubDate.last10.log, S.T.state, S.T.one, S.npnct04.log, S.T.share, PubDate.hour.fctr, H.T.report, PubDate.second.fctr, H.T.newyork, S.T.can, H.T.say, A.T.week, H.T.fashion, H.npnct16.log, A.T.will, S.T.show, PubDate.minute.fctr, H.ndgts.log, S.npnct12.log, H.npnct07.log, A.nwrds.log, H.T.art, H.npnct01.log, .rnorm, S.T.time, H.npnct13.log, S.npnct01.log, A.T.intern, A.T.compani, H.T.new, A.nchrs.log, PubDate.last1.log, H.npnct14.log, H.T.first, PubDate.date.fctr, H.T.busi, S.npnct13.log, H.T.day, H.T.china, A.T.year, H.nwrds.log, PubDate.wkend, H.T.big, H.T.news, H.T.week, H.has.ebola, H.T.take, H.T.bank, S.T.take, S.T.day, H.T.make, H.T.time, H.T.pictur, H.T.billion, H.T.X2014, A.T.articl, H.T.obama, S.npnct16.log, PubDate.last100.log, S.npnct15.log, A.T.first, H.T.test, A.npnct17.log, S.npnct06.log, S.T.new, H.npnct30.log, A.T.fashion, H.T.springsumm, H.T.deal, S.npnct30.log, H.npnct02.log, S.npnct22.log, S.npnct07.log, A.T.presid, S.T.presid, S.npnct03.log, S.has.year.colon, H.npnct05.log, S.npnct08.log, S.npnct09.log, A.npnct19.log, A.npnct20.log"
## Aggregating results
## Fitting final model on full training set## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning: not plotting observations with leverage one:
## 3675
## Warning: not plotting observations with leverage one:
## 3675
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.8589 -0.3312 -0.1521 -0.0001 3.5098
##
## Coefficients:
## Estimate
## (Intercept) -4.580e+00
## WordCount.log 1.132e+00
## `myCategory.fctrForeign#World#Asia Pacific` -4.383e+00
## `myCategory.fctr#Multimedia#` -4.536e+00
## `myCategory.fctrCulture#Arts#` -2.583e+00
## `myCategory.fctrBusiness#Business Day#Dealbook` -2.461e+00
## myCategory.fctrmyOther -1.983e+01
## `myCategory.fctrBusiness#Technology#` -1.865e+00
## `myCategory.fctrBusiness#Crosswords/Games#` 6.843e-01
## `myCategory.fctrTStyle##` -4.284e+00
## `myCategory.fctrForeign#World#` -1.847e+01
## `myCategory.fctrOpEd#Opinion#` 8.945e-01
## `myCategory.fctrStyles##Fashion` -5.215e+00
## `myCategory.fctr#Opinion#Room For Debate` -5.508e+00
## `myCategory.fctr#U.S.#Education` -2.040e+01
## `myCategory.fctr##` -2.532e+00
## `myCategory.fctrMetro#N.Y. / Region#` -1.780e+00
## `myCategory.fctrBusiness#Business Day#Small Business` -3.982e+00
## `myCategory.fctrStyles#U.S.#` -3.955e-01
## `myCategory.fctrTravel#Travel#` -4.279e+00
## `myCategory.fctr#Opinion#The Public Editor` 4.257e-01
## H.npnct21.log 1.422e+00
## A.npnct21.log 1.455e+00
## S.nuppr.log -6.036e-01
## H.nchrs.log -1.331e+00
## A.npnct14.log 8.802e-01
## H.npnct09.log 1.831e+00
## PubDate.wkday.fctr1 -5.528e-01
## PubDate.wkday.fctr2 -9.933e-01
## PubDate.wkday.fctr3 -7.135e-01
## PubDate.wkday.fctr4 -8.242e-01
## PubDate.wkday.fctr5 -8.435e-01
## PubDate.wkday.fctr6 -8.851e-01
## H.nuppr.log 1.034e+00
## H.T.polit -6.893e-01
## A.T.newyork 1.742e+00
## A.T.report -1.452e+00
## H.npnct12.log 4.047e-01
## S.ndgts.log -2.679e-01
## A.nwrds.unq.log -1.020e+00
## S.T.said 1.887e+00
## S.T.make -6.393e-01
## H.npnct17.log 5.954e-01
## PubDate.last10.log 1.767e-01
## S.T.state 7.804e-01
## S.T.one -4.086e-01
## S.npnct04.log -9.166e-01
## S.T.share -1.514e+00
## `PubDate.hour.fctr(7.67,15.3]` 1.402e-01
## `PubDate.hour.fctr(15.3,23]` 2.477e-01
## H.T.report -7.371e-01
## `PubDate.second.fctr(14.8,29.5]` -6.471e-02
## `PubDate.second.fctr(29.5,44.2]` -2.140e-02
## `PubDate.second.fctr(44.2,59.1]` -2.151e-01
## H.T.newyork -1.170e+00
## S.T.can -1.164e+00
## H.T.say -4.180e-01
## A.T.week -1.019e+00
## H.T.fashion 6.703e-02
## H.npnct16.log -1.275e-01
## A.T.will -9.116e-01
## S.T.show -6.261e-01
## `PubDate.minute.fctr(14.8,29.5]` -1.688e-02
## `PubDate.minute.fctr(29.5,44.2]` -2.180e-01
## `PubDate.minute.fctr(44.2,59.1]` 1.130e-02
## H.ndgts.log 8.880e-02
## S.npnct12.log -1.259e-01
## H.npnct07.log 8.607e-02
## A.nwrds.log -2.897e-01
## H.T.art -3.084e-01
## H.npnct01.log -1.485e+00
## .rnorm -6.598e-02
## S.T.time -1.106e+00
## H.npnct13.log 1.243e-01
## S.npnct01.log 1.983e+00
## A.T.intern 8.152e-01
## A.T.compani -1.002e+00
## H.T.new -4.218e-01
## A.nchrs.log 5.151e-01
## PubDate.last1.log -3.106e-02
## H.npnct14.log -1.347e-01
## H.T.first -2.353e-01
## `PubDate.date.fctr(7,13]` 8.340e-02
## `PubDate.date.fctr(13,19]` -1.227e-01
## `PubDate.date.fctr(19,25]` -8.868e-03
## `PubDate.date.fctr(25,31]` 4.946e-02
## H.T.busi -1.381e+00
## S.npnct13.log -1.761e-01
## H.T.day -3.946e-01
## H.T.china -1.417e-01
## A.T.year -1.011e+00
## H.nwrds.log 5.816e-01
## PubDate.wkend -1.137e-01
## H.T.big -1.492e-01
## H.T.news -7.471e-01
## H.T.week -3.992e-01
## H.has.ebola -4.503e-01
## H.T.take -2.586e-01
## H.T.bank 2.024e-01
## S.T.take -1.587e-01
## S.T.day -1.595e-01
## H.T.make -2.822e-01
## H.T.time -6.797e-02
## H.T.pictur 3.020e-01
## H.T.billion 3.767e-01
## H.T.X2014 -2.767e-01
## A.T.articl 7.530e-01
## H.T.obama 6.605e-02
## S.npnct16.log -3.190e-01
## PubDate.last100.log -9.819e-03
## S.npnct15.log 3.955e-01
## A.T.first -2.251e-01
## H.T.test 1.683e-01
## A.npnct17.log 2.839e-01
## S.npnct06.log 1.008e+00
## S.T.new -3.070e-01
## H.npnct30.log 4.750e-02
## A.T.fashion -2.764e+00
## H.T.springsumm -1.383e+01
## H.T.deal -2.460e-01
## S.npnct30.log -1.464e+01
## H.npnct02.log -1.673e+01
## S.npnct22.log -2.288e+01
## S.npnct07.log -2.482e+01
## A.T.presid 7.504e+02
## S.T.presid -7.504e+02
## S.npnct03.log -2.826e+01
## S.has.year.colon -1.251e+01
## H.npnct05.log 8.674e-01
## S.npnct08.log 1.212e+01
## S.npnct09.log -1.182e+01
## A.npnct19.log 1.175e+00
## A.npnct20.log -1.739e+01
## Std. Error z value
## (Intercept) 2.278e+00 -2.011
## WordCount.log 7.303e-02 15.495
## `myCategory.fctrForeign#World#Asia Pacific` 6.541e-01 -6.701
## `myCategory.fctr#Multimedia#` 7.722e-01 -5.874
## `myCategory.fctrCulture#Arts#` 2.912e-01 -8.869
## `myCategory.fctrBusiness#Business Day#Dealbook` 2.532e-01 -9.719
## myCategory.fctrmyOther 1.545e+03 -0.013
## `myCategory.fctrBusiness#Technology#` 2.632e-01 -7.087
## `myCategory.fctrBusiness#Crosswords/Games#` 3.790e-01 1.806
## `myCategory.fctrTStyle##` 4.127e-01 -10.381
## `myCategory.fctrForeign#World#` 7.558e+02 -0.024
## `myCategory.fctrOpEd#Opinion#` 2.432e-01 3.678
## `myCategory.fctrStyles##Fashion` 1.250e+00 -4.173
## `myCategory.fctr#Opinion#Room For Debate` 5.237e-01 -10.517
## `myCategory.fctr#U.S.#Education` 5.075e+02 -0.040
## `myCategory.fctr##` 2.321e-01 -10.910
## `myCategory.fctrMetro#N.Y. / Region#` 4.154e-01 -4.284
## `myCategory.fctrBusiness#Business Day#Small Business` 5.446e-01 -7.312
## `myCategory.fctrStyles#U.S.#` 2.715e-01 -1.457
## `myCategory.fctrTravel#Travel#` 1.030e+00 -4.156
## `myCategory.fctr#Opinion#The Public Editor` 6.765e-01 0.629
## H.npnct21.log 2.571e-01 5.530
## A.npnct21.log 2.676e-01 5.438
## S.nuppr.log 1.275e-01 -4.735
## H.nchrs.log 2.897e-01 -4.593
## A.npnct14.log 2.126e-01 4.141
## H.npnct09.log 6.200e-01 2.953
## PubDate.wkday.fctr1 4.234e-01 -1.306
## PubDate.wkday.fctr2 4.620e-01 -2.150
## PubDate.wkday.fctr3 4.568e-01 -1.562
## PubDate.wkday.fctr4 4.503e-01 -1.830
## PubDate.wkday.fctr5 4.565e-01 -1.848
## PubDate.wkday.fctr6 3.773e-01 -2.346
## H.nuppr.log 3.367e-01 3.072
## H.T.polit 2.340e-01 -2.945
## A.T.newyork 8.710e-01 2.000
## A.T.report 8.333e-01 -1.742
## H.npnct12.log 1.704e-01 2.375
## S.ndgts.log 1.216e-01 -2.203
## A.nwrds.unq.log 4.446e-01 -2.295
## S.T.said 6.490e-01 2.907
## S.T.make 4.942e-01 -1.294
## H.npnct17.log 4.540e-01 1.312
## PubDate.last10.log 9.553e-02 1.850
## S.T.state 6.689e-01 1.167
## S.T.one 4.931e-01 -0.829
## S.npnct04.log 5.052e-01 -1.815
## S.T.share 8.779e-01 -1.725
## `PubDate.hour.fctr(7.67,15.3]` 1.932e-01 0.726
## `PubDate.hour.fctr(15.3,23]` 1.945e-01 1.273
## H.T.report 5.307e-01 -1.389
## `PubDate.second.fctr(14.8,29.5]` 1.417e-01 -0.457
## `PubDate.second.fctr(29.5,44.2]` 1.402e-01 -0.153
## `PubDate.second.fctr(44.2,59.1]` 1.425e-01 -1.509
## H.T.newyork 4.583e-01 -2.554
## S.T.can 6.028e-01 -1.931
## H.T.say 3.577e-01 -1.169
## A.T.week 7.170e-01 -1.421
## H.T.fashion 1.380e+00 0.049
## H.npnct16.log 2.256e-01 -0.565
## A.T.will 6.088e-01 -1.497
## S.T.show 7.371e-01 -0.849
## `PubDate.minute.fctr(14.8,29.5]` 1.457e-01 -0.116
## `PubDate.minute.fctr(29.5,44.2]` 1.425e-01 -1.529
## `PubDate.minute.fctr(44.2,59.1]` 1.481e-01 0.076
## H.ndgts.log 1.899e-01 0.468
## S.npnct12.log 1.165e-01 -1.080
## H.npnct07.log 1.517e-01 0.567
## A.nwrds.log 6.259e-01 -0.463
## H.T.art 5.938e-01 -0.519
## H.npnct01.log 9.421e-01 -1.577
## .rnorm 5.057e-02 -1.305
## S.T.time 7.164e-01 -1.543
## H.npnct13.log 2.485e-01 0.500
## S.npnct01.log 1.115e+00 1.778
## A.T.intern 1.416e+00 0.576
## A.T.compani 7.134e-01 -1.404
## H.T.new 3.648e-01 -1.156
## A.nchrs.log 3.980e-01 1.294
## PubDate.last1.log 3.563e-02 -0.872
## H.npnct14.log 1.622e-01 -0.830
## H.T.first 6.099e-01 -0.386
## `PubDate.date.fctr(7,13]` 1.578e-01 0.528
## `PubDate.date.fctr(13,19]` 1.571e-01 -0.781
## `PubDate.date.fctr(19,25]` 1.541e-01 -0.058
## `PubDate.date.fctr(25,31]` 1.672e-01 0.296
## H.T.busi 8.079e-01 -1.709
## S.npnct13.log 1.628e-01 -1.082
## H.T.day 4.379e-01 -0.901
## H.T.china 5.662e-01 -0.250
## A.T.year 7.575e-01 -1.335
## H.nwrds.log 4.800e-01 1.212
## PubDate.wkend 3.566e-01 -0.319
## H.T.big 3.870e-01 -0.385
## H.T.news 7.918e-01 -0.944
## H.T.week 4.619e-01 -0.864
## H.has.ebola 3.747e-01 -1.202
## H.T.take 3.182e-01 -0.813
## H.T.bank 3.689e-01 0.549
## S.T.take 7.679e-01 -0.207
## S.T.day 8.141e-01 -0.196
## H.T.make 2.846e-01 -0.992
## H.T.time 2.821e-01 -0.241
## H.T.pictur 6.187e-01 0.488
## H.T.billion 5.157e-01 0.730
## H.T.X2014 8.453e-01 -0.327
## A.T.articl 1.423e+00 0.529
## H.T.obama 3.993e-01 0.165
## S.npnct16.log 4.026e-01 -0.792
## PubDate.last100.log 3.571e-02 -0.275
## S.npnct15.log 1.421e+00 0.278
## A.T.first 8.770e-01 -0.257
## H.T.test 5.353e-01 0.314
## A.npnct17.log 9.910e-01 0.287
## S.npnct06.log 8.192e-01 1.231
## S.T.new 6.132e-01 -0.501
## H.npnct30.log 1.612e+00 0.029
## A.T.fashion 2.789e+00 -0.991
## H.T.springsumm 9.107e+02 -0.015
## H.T.deal 4.652e-01 -0.529
## S.npnct30.log 1.023e+03 -0.014
## H.npnct02.log 2.128e+03 -0.008
## S.npnct22.log 3.813e+03 -0.006
## S.npnct07.log 5.528e+03 -0.004
## A.T.presid 1.014e+05 0.007
## S.T.presid 1.014e+05 -0.007
## S.npnct03.log 5.324e+03 -0.005
## S.has.year.colon 2.790e+03 -0.004
## H.npnct05.log 1.602e+00 0.541
## S.npnct08.log 7.757e+03 0.002
## S.npnct09.log 7.757e+03 -0.002
## A.npnct19.log 3.582e+04 0.000
## A.npnct20.log 2.342e+04 -0.001
## Pr(>|z|)
## (Intercept) 0.044360 *
## WordCount.log < 2e-16 ***
## `myCategory.fctrForeign#World#Asia Pacific` 2.07e-11 ***
## `myCategory.fctr#Multimedia#` 4.26e-09 ***
## `myCategory.fctrCulture#Arts#` < 2e-16 ***
## `myCategory.fctrBusiness#Business Day#Dealbook` < 2e-16 ***
## myCategory.fctrmyOther 0.989757
## `myCategory.fctrBusiness#Technology#` 1.37e-12 ***
## `myCategory.fctrBusiness#Crosswords/Games#` 0.070966 .
## `myCategory.fctrTStyle##` < 2e-16 ***
## `myCategory.fctrForeign#World#` 0.980500
## `myCategory.fctrOpEd#Opinion#` 0.000235 ***
## `myCategory.fctrStyles##Fashion` 3.01e-05 ***
## `myCategory.fctr#Opinion#Room For Debate` < 2e-16 ***
## `myCategory.fctr#U.S.#Education` 0.967931
## `myCategory.fctr##` < 2e-16 ***
## `myCategory.fctrMetro#N.Y. / Region#` 1.84e-05 ***
## `myCategory.fctrBusiness#Business Day#Small Business` 2.63e-13 ***
## `myCategory.fctrStyles#U.S.#` 0.145175
## `myCategory.fctrTravel#Travel#` 3.24e-05 ***
## `myCategory.fctr#Opinion#The Public Editor` 0.529129
## H.npnct21.log 3.20e-08 ***
## A.npnct21.log 5.38e-08 ***
## S.nuppr.log 2.19e-06 ***
## H.nchrs.log 4.37e-06 ***
## A.npnct14.log 3.46e-05 ***
## H.npnct09.log 0.003150 **
## PubDate.wkday.fctr1 0.191689
## PubDate.wkday.fctr2 0.031529 *
## PubDate.wkday.fctr3 0.118263
## PubDate.wkday.fctr4 0.067189 .
## PubDate.wkday.fctr5 0.064644 .
## PubDate.wkday.fctr6 0.018985 *
## H.nuppr.log 0.002128 **
## H.T.polit 0.003225 **
## A.T.newyork 0.045487 *
## A.T.report 0.081485 .
## H.npnct12.log 0.017558 *
## S.ndgts.log 0.027580 *
## A.nwrds.unq.log 0.021745 *
## S.T.said 0.003649 **
## S.T.make 0.195802
## H.npnct17.log 0.189650
## PubDate.last10.log 0.064320 .
## S.T.state 0.243353
## S.T.one 0.407304
## S.npnct04.log 0.069599 .
## S.T.share 0.084547 .
## `PubDate.hour.fctr(7.67,15.3]` 0.468123
## `PubDate.hour.fctr(15.3,23]` 0.202887
## H.T.report 0.164811
## `PubDate.second.fctr(14.8,29.5]` 0.648007
## `PubDate.second.fctr(29.5,44.2]` 0.878694
## `PubDate.second.fctr(44.2,59.1]` 0.131303
## H.T.newyork 0.010658 *
## S.T.can 0.053520 .
## H.T.say 0.242527
## A.T.week 0.155314
## H.T.fashion 0.961274
## H.npnct16.log 0.572028
## A.T.will 0.134278
## S.T.show 0.395654
## `PubDate.minute.fctr(14.8,29.5]` 0.907813
## `PubDate.minute.fctr(29.5,44.2]` 0.126161
## `PubDate.minute.fctr(44.2,59.1]` 0.939151
## H.ndgts.log 0.640058
## S.npnct12.log 0.280080
## H.npnct07.log 0.570398
## A.nwrds.log 0.643493
## H.T.art 0.603560
## H.npnct01.log 0.114898
## .rnorm 0.192019
## S.T.time 0.122746
## H.npnct13.log 0.616952
## S.npnct01.log 0.075468 .
## A.T.intern 0.564947
## A.T.compani 0.160353
## H.T.new 0.247523
## A.nchrs.log 0.195622
## PubDate.last1.log 0.383264
## H.npnct14.log 0.406344
## H.T.first 0.699672
## `PubDate.date.fctr(7,13]` 0.597228
## `PubDate.date.fctr(13,19]` 0.434782
## `PubDate.date.fctr(19,25]` 0.954122
## `PubDate.date.fctr(25,31]` 0.767377
## H.T.busi 0.087482 .
## S.npnct13.log 0.279361
## H.T.day 0.367460
## H.T.china 0.802376
## A.T.year 0.181786
## H.nwrds.log 0.225616
## PubDate.wkend 0.749792
## H.T.big 0.699876
## H.T.news 0.345413
## H.T.week 0.387429
## H.has.ebola 0.229409
## H.T.take 0.416465
## H.T.bank 0.583244
## S.T.take 0.836316
## S.T.day 0.844683
## H.T.make 0.321396
## H.T.time 0.809616
## H.T.pictur 0.625435
## H.T.billion 0.465133
## H.T.X2014 0.743454
## A.T.articl 0.596686
## H.T.obama 0.868617
## S.npnct16.log 0.428243
## PubDate.last100.log 0.783367
## S.npnct15.log 0.780731
## A.T.first 0.797450
## H.T.test 0.753147
## A.npnct17.log 0.774489
## S.npnct06.log 0.218501
## S.T.new 0.616655
## H.npnct30.log 0.976485
## A.T.fashion 0.321664
## H.T.springsumm 0.987887
## H.T.deal 0.596925
## S.npnct30.log 0.988587
## H.npnct02.log 0.993728
## S.npnct22.log 0.995212
## S.npnct07.log 0.996418
## A.T.presid 0.994093
## S.T.presid 0.994093
## S.npnct03.log 0.995766
## S.has.year.colon 0.996422
## H.npnct05.log 0.588330
## S.npnct08.log 0.998753
## S.npnct09.log 0.998785
## A.npnct19.log 0.999974
## A.npnct20.log 0.999407
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 5900.1 on 6531 degrees of freedom
## Residual deviance: 2779.0 on 6399 degrees of freedom
## AIC: 3045
##
## Number of Fisher Scoring iterations: 18
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.2866885
## 2 0.1 0.6578266
## 3 0.2 0.7322619
## 4 0.3 0.7425870
## 5 0.4 0.7387305
## 6 0.5 0.7285291
## 7 0.6 0.6933614
## 8 0.7 0.6497748
## 9 0.8 0.5554859
## 10 0.9 0.3646973
## 11 1.0 0.0000000## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## Popular.fctr Popular.fctr.predict.Final.glm.N
## 1 N 5069
## 2 Y 229
## Popular.fctr.predict.Final.glm.Y
## 1 370
## 2 864
## Prediction
## Reference N Y
## N 5069 370
## Y 229 864
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 9.082976e-01 6.870476e-01 9.010386e-01 9.151896e-01 8.326699e-01
## AccuracyPValue McnemarPValue
## 9.726144e-70 1.063706e-08## Warning in mypredict_mdl(mdl, df = fit_df, rsp_var, rsp_var_out,
## model_id_method, : Expecting 1 metric: Accuracy; recd: Accuracy, Kappa;
## retaining Accuracy only
## model_id model_method
## 1 Final.glm glm
## feats
## 1 WordCount.log, myCategory.fctr, H.npnct21.log, A.npnct21.log, S.nuppr.log, H.nchrs.log, A.npnct14.log, H.npnct09.log, PubDate.wkday.fctr, H.nuppr.log, H.T.polit, A.T.newyork, A.T.report, H.npnct12.log, S.ndgts.log, A.nwrds.unq.log, S.T.said, S.T.make, H.npnct17.log, PubDate.last10.log, S.T.state, S.T.one, S.npnct04.log, S.T.share, PubDate.hour.fctr, H.T.report, PubDate.second.fctr, H.T.newyork, S.T.can, H.T.say, A.T.week, H.T.fashion, H.npnct16.log, A.T.will, S.T.show, PubDate.minute.fctr, H.ndgts.log, S.npnct12.log, H.npnct07.log, A.nwrds.log, H.T.art, H.npnct01.log, .rnorm, S.T.time, H.npnct13.log, S.npnct01.log, A.T.intern, A.T.compani, H.T.new, A.nchrs.log, PubDate.last1.log, H.npnct14.log, H.T.first, PubDate.date.fctr, H.T.busi, S.npnct13.log, H.T.day, H.T.china, A.T.year, H.nwrds.log, PubDate.wkend, H.T.big, H.T.news, H.T.week, H.has.ebola, H.T.take, H.T.bank, S.T.take, S.T.day, H.T.make, H.T.time, H.T.pictur, H.T.billion, H.T.X2014, A.T.articl, H.T.obama, S.npnct16.log, PubDate.last100.log, S.npnct15.log, A.T.first, H.T.test, A.npnct17.log, S.npnct06.log, S.T.new, H.npnct30.log, A.T.fashion, H.T.springsumm, H.T.deal, S.npnct30.log, H.npnct02.log, S.npnct22.log, S.npnct07.log, A.T.presid, S.T.presid, S.npnct03.log, S.has.year.colon, H.npnct05.log, S.npnct08.log, S.npnct09.log, A.npnct19.log, A.npnct20.log
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 11.575 5.408
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.944687 0.3 0.742587 0.9070715
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit min.aic.fit
## 1 0.9010386 0.9151896 0.6423119 3045.02
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.01164752 0.05110743glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.data.training", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 14 fit.data.training 8 0 335.359 354.08 18.721
## 15 fit.data.training 8 1 354.080 NA NAglb_trnent_df <- glb_get_predictions(df=glb_trnent_df, mdl_id=glb_fin_mdl_id,
rsp_var_out=glb_rsp_var_out,
prob_threshold_def=ifelse(glb_is_classification && glb_is_binomial,
glb_models_df[glb_models_df$model_id == glb_sel_mdl_id, "opt.prob.threshold.OOB"], NULL))## Warning in glb_get_predictions(df = glb_trnent_df, mdl_id =
## glb_fin_mdl_id, : Using default probability threshold: 0.3glb_feats_df <- mymerge_feats_importance(feats_df=glb_feats_df, sel_mdl=glb_fin_mdl,
entity_df=glb_trnent_df)
glb_feats_df[, paste0(glb_fin_mdl_id, ".importance")] <- glb_feats_df$importance
print(glb_feats_df)## id importance cor.y
## WordCount.log WordCount.log 1.000000e+02 2.656836e-01
## myCategory.fctr myCategory.fctr 7.312143e+01 1.234541e-02
## H.npnct21.log H.npnct21.log 3.777302e+01 1.283641e-01
## A.npnct21.log A.npnct21.log 3.475205e+01 5.482747e-02
## S.nuppr.log S.nuppr.log 3.238555e+01 -2.718459e-01
## H.nchrs.log H.nchrs.log 3.201395e+01 -1.710624e-01
## A.npnct14.log A.npnct14.log 2.764418e+01 -4.999563e-02
## H.npnct09.log H.npnct09.log 2.188969e+01 5.375262e-02
## PubDate.wkday.fctr PubDate.wkday.fctr 2.165613e+01 -3.980129e-02
## H.nuppr.log H.nuppr.log 2.131190e+01 -1.278085e-01
## H.T.polit H.T.polit 2.052535e+01 -3.062866e-02
## A.T.newyork A.T.newyork 1.961261e+01 -5.706083e-02
## A.T.report A.T.report 1.938341e+01 -4.741555e-02
## H.npnct12.log H.npnct12.log 1.796841e+01 1.333613e-02
## S.ndgts.log S.ndgts.log 1.621363e+01 -1.242046e-01
## A.nwrds.unq.log A.nwrds.unq.log 1.619197e+01 -2.506012e-01
## S.T.said S.T.said 1.520654e+01 1.826884e-02
## S.T.make S.T.make 1.512129e+01 3.959645e-02
## H.npnct17.log H.npnct17.log 1.480733e+01 3.039622e-02
## PubDate.last10.log PubDate.last10.log 1.453274e+01 4.931702e-02
## S.T.state S.T.state 1.414282e+01 1.012205e-02
## S.T.one S.T.one 1.375899e+01 1.080534e-02
## S.npnct04.log S.npnct04.log 1.373898e+01 -6.294642e-02
## S.T.share S.T.share 1.314940e+01 -5.070234e-02
## PubDate.hour.fctr PubDate.hour.fctr 1.314384e+01 1.354368e-01
## H.T.report H.T.report 1.299102e+01 -6.244050e-02
## PubDate.second.fctr PubDate.second.fctr 1.273156e+01 -1.187946e-02
## H.T.newyork H.T.newyork 1.269274e+01 -5.650839e-02
## S.T.can S.T.can 1.228026e+01 3.049697e-02
## H.T.say H.T.say 1.222772e+01 -9.960773e-03
## A.T.week A.T.week 1.168514e+01 -8.492895e-02
## H.T.fashion H.T.fashion 1.151035e+01 -7.947505e-02
## H.npnct16.log H.npnct16.log 1.136051e+01 -8.273237e-02
## A.T.will A.T.will 1.129252e+01 -3.887937e-02
## S.T.show S.T.show 1.089884e+01 -4.193803e-02
## PubDate.minute.fctr PubDate.minute.fctr 1.041575e+01 -3.407385e-02
## H.ndgts.log H.ndgts.log 1.035074e+01 -1.196633e-01
## S.npnct12.log S.npnct12.log 9.930532e+00 -9.158156e-02
## H.npnct07.log H.npnct07.log 9.854654e+00 -1.201741e-02
## A.nwrds.log A.nwrds.log 9.770236e+00 1.354108e-01
## H.T.art H.T.art 9.623245e+00 -3.280483e-02
## H.npnct01.log H.npnct01.log 9.509217e+00 2.271577e-02
## .rnorm .rnorm 9.238260e+00 -8.244230e-03
## S.T.time S.T.time 9.237301e+00 -5.303654e-02
## H.npnct13.log H.npnct13.log 9.109144e+00 -1.305305e-02
## S.npnct01.log S.npnct01.log 8.749827e+00 3.093101e-02
## A.T.intern A.T.intern 8.472657e+00 -6.949870e-02
## A.T.compani A.T.compani 8.369682e+00 -4.751471e-02
## H.T.new H.T.new 8.270503e+00 -4.327803e-02
## A.nchrs.log A.nchrs.log 8.248780e+00 -2.245488e-01
## PubDate.last1.log PubDate.last1.log 8.146784e+00 4.635751e-02
## H.npnct14.log H.npnct14.log 8.031911e+00 -2.524770e-02
## H.T.first H.T.first 7.798393e+00 -4.458885e-02
## PubDate.date.fctr PubDate.date.fctr 7.300560e+00 -1.164756e-02
## H.T.busi H.T.busi 7.168757e+00 -4.901905e-02
## S.npnct13.log S.npnct13.log 6.710517e+00 -3.638891e-02
## H.T.day H.T.day 6.378120e+00 -6.033488e-02
## H.T.china H.T.china 6.350450e+00 -3.144808e-02
## A.T.year A.T.year 6.178982e+00 -4.721236e-02
## H.nwrds.log H.nwrds.log 6.165115e+00 1.410282e-01
## PubDate.wkend PubDate.wkend 6.020520e+00 1.067288e-01
## H.T.big H.T.big 6.007063e+00 -1.390748e-02
## H.T.news H.T.news 5.524332e+00 -4.415284e-02
## H.T.week H.T.week 5.347019e+00 -6.812724e-02
## H.has.ebola H.has.ebola 4.853552e+00 2.588140e-02
## H.T.take H.T.take 4.592147e+00 -8.582583e-04
## H.T.bank H.T.bank 4.574012e+00 -9.989139e-03
## S.T.take S.T.take 4.545539e+00 -2.275732e-02
## S.T.day S.T.day 4.501029e+00 -4.188671e-02
## H.T.make H.T.make 4.025030e+00 1.430572e-02
## H.T.time H.T.time 3.968662e+00 -2.527450e-03
## H.T.pictur H.T.pictur 3.751942e+00 -3.993172e-02
## H.T.billion H.T.billion 3.746828e+00 -2.949817e-02
## H.T.X2014 H.T.X2014 2.938640e+00 -4.497745e-02
## A.T.articl A.T.articl 2.744290e+00 -5.445243e-02
## H.T.obama H.T.obama 2.631610e+00 -9.907543e-03
## S.npnct16.log S.npnct16.log 2.586897e+00 -6.770952e-02
## PubDate.last100.log PubDate.last100.log 2.562314e+00 -7.663322e-03
## S.npnct15.log S.npnct15.log 1.951937e+00 -2.121844e-02
## A.T.first A.T.first 1.686309e+00 -4.603341e-02
## H.T.test H.T.test 1.393050e+00 -2.065255e-02
## A.npnct17.log A.npnct17.log 1.012142e+00 -1.587454e-03
## S.npnct06.log S.npnct06.log 7.949129e-01 -2.389145e-02
## S.T.new S.T.new 6.136717e-01 -2.769558e-02
## H.npnct30.log H.npnct30.log 5.296067e-01 -8.917338e-02
## A.T.fashion A.T.fashion 1.689321e-01 -8.419345e-02
## H.T.springsumm H.T.springsumm 8.668136e-02 -5.943248e-02
## H.T.deal H.T.deal 6.560495e-02 -2.559418e-02
## S.npnct30.log S.npnct30.log 5.692475e-02 -4.370037e-02
## H.npnct02.log H.npnct02.log 2.826878e-02 -2.001851e-02
## S.npnct22.log S.npnct22.log 2.688801e-02 -1.923169e-02
## S.npnct07.log S.npnct07.log 2.677639e-02 -1.214357e-02
## A.T.presid A.T.presid 2.652248e-02 -1.789086e-03
## S.T.presid S.T.presid 2.649780e-02 -2.079562e-03
## S.npnct03.log S.npnct03.log 2.562697e-02 -1.240734e-02
## S.has.year.colon S.has.year.colon 2.044323e-02 -1.755336e-02
## H.npnct05.log H.npnct05.log 1.884000e-02 -9.653967e-03
## S.npnct08.log S.npnct08.log 7.744037e-03 -2.413868e-03
## S.npnct09.log S.npnct09.log 6.828413e-03 -3.986882e-03
## A.npnct19.log A.npnct19.log 3.032825e-03 -1.271661e-02
## A.npnct20.log A.npnct20.log 0.000000e+00 -1.451467e-02
## A.has.http A.has.http NA -1.359260e-02
## A.has.year.colon A.has.year.colon NA -1.755336e-02
## A.ndgts.log A.ndgts.log NA -1.249484e-01
## A.npnct01.log A.npnct01.log NA 3.093101e-02
## A.npnct02.log A.npnct02.log NA -1.451467e-02
## A.npnct03.log A.npnct03.log NA -1.359260e-02
## A.npnct04.log A.npnct04.log NA -6.294642e-02
## A.npnct05.log A.npnct05.log NA NA
## A.npnct06.log A.npnct06.log NA -2.389145e-02
## A.npnct07.log A.npnct07.log NA -1.214357e-02
## A.npnct08.log A.npnct08.log NA -3.258100e-03
## A.npnct09.log A.npnct09.log NA -4.775988e-03
## A.npnct10.log A.npnct10.log NA NA
## A.npnct11.log A.npnct11.log NA -5.547032e-03
## A.npnct12.log A.npnct12.log NA -9.183870e-02
## A.npnct13.log A.npnct13.log NA -3.760012e-02
## A.npnct15.log A.npnct15.log NA -2.407715e-02
## A.npnct16.log A.npnct16.log NA -6.893301e-02
## A.npnct18.log A.npnct18.log NA -1.451467e-02
## A.npnct22.log A.npnct22.log NA -1.923169e-02
## A.npnct23.log A.npnct23.log NA 1.537569e-02
## A.npnct24.log A.npnct24.log NA NA
## A.npnct25.log A.npnct25.log NA 1.537569e-02
## A.npnct26.log A.npnct26.log NA -9.890046e-19
## A.npnct27.log A.npnct27.log NA -5.547032e-03
## A.npnct28.log A.npnct28.log NA NA
## A.npnct29.log A.npnct29.log NA NA
## A.npnct30.log A.npnct30.log NA -4.373349e-02
## A.npnct31.log A.npnct31.log NA NA
## A.npnct32.log A.npnct32.log NA NA
## A.nuppr.log A.nuppr.log NA -2.720962e-01
## A.T.can A.T.can NA 3.127063e-02
## A.T.day A.T.day NA -4.196599e-02
## A.T.make A.T.make NA 3.965722e-02
## A.T.new A.T.new NA -2.782876e-02
## A.T.one A.T.one NA 1.081694e-02
## A.T.said A.T.said NA 1.839871e-02
## A.T.share A.T.share NA -5.070234e-02
## A.T.show A.T.show NA -4.196129e-02
## A.T.state A.T.state NA 1.020706e-02
## A.T.take A.T.take NA -2.282555e-02
## A.T.time A.T.time NA -5.313395e-02
## clusterid clusterid NA NA
## H.has.http H.has.http NA NA
## H.has.year.colon H.has.year.colon NA -7.842875e-02
## H.npnct03.log H.npnct03.log NA 9.533020e-03
## H.npnct04.log H.npnct04.log NA -5.126277e-02
## H.npnct06.log H.npnct06.log NA 3.190718e-02
## H.npnct08.log H.npnct08.log NA 5.375262e-02
## H.npnct10.log H.npnct10.log NA NA
## H.npnct11.log H.npnct11.log NA -5.547032e-03
## H.npnct15.log H.npnct15.log NA -6.158577e-02
## H.npnct18.log H.npnct18.log NA NA
## H.npnct19.log H.npnct19.log NA NA
## H.npnct20.log H.npnct20.log NA NA
## H.npnct22.log H.npnct22.log NA -5.547032e-03
## H.npnct23.log H.npnct23.log NA NA
## H.npnct24.log H.npnct24.log NA NA
## H.npnct25.log H.npnct25.log NA NA
## H.npnct26.log H.npnct26.log NA -9.890046e-19
## H.npnct27.log H.npnct27.log NA NA
## H.npnct28.log H.npnct28.log NA NA
## H.npnct29.log H.npnct29.log NA NA
## H.npnct31.log H.npnct31.log NA NA
## H.npnct32.log H.npnct32.log NA NA
## H.nwrds.unq.log H.nwrds.unq.log NA -2.044964e-01
## H.T.daili H.T.daili NA -6.299948e-02
## H.T.morn H.T.morn NA -4.838380e-02
## H.T.today H.T.today NA -5.831308e-02
## H.T.X2015 H.T.X2015 NA -6.601141e-02
## Popular Popular NA 1.000000e+00
## Popular.fctr Popular.fctr NA NA
## PubDate.last1 PubDate.last1 NA 3.592267e-02
## PubDate.last10 PubDate.last10 NA 5.398093e-02
## PubDate.last100 PubDate.last100 NA 3.989229e-02
## PubDate.month.fctr PubDate.month.fctr NA 1.914874e-02
## PubDate.POSIX PubDate.POSIX NA 1.568326e-02
## PubDate.year.fctr PubDate.year.fctr NA NA
## PubDate.zoo PubDate.zoo NA 1.568326e-02
## S.has.http S.has.http NA NA
## S.nchrs.log S.nchrs.log NA -2.246930e-01
## S.npnct02.log S.npnct02.log NA -5.547032e-03
## S.npnct05.log S.npnct05.log NA NA
## S.npnct10.log S.npnct10.log NA NA
## S.npnct11.log S.npnct11.log NA -5.547032e-03
## S.npnct14.log S.npnct14.log NA -5.332519e-02
## S.npnct17.log S.npnct17.log NA -1.587454e-03
## S.npnct18.log S.npnct18.log NA NA
## S.npnct19.log S.npnct19.log NA NA
## S.npnct20.log S.npnct20.log NA NA
## S.npnct21.log S.npnct21.log NA 5.503894e-02
## S.npnct23.log S.npnct23.log NA 2.760321e-02
## S.npnct24.log S.npnct24.log NA NA
## S.npnct25.log S.npnct25.log NA 2.760321e-02
## S.npnct26.log S.npnct26.log NA -9.890046e-19
## S.npnct27.log S.npnct27.log NA NA
## S.npnct28.log S.npnct28.log NA NA
## S.npnct29.log S.npnct29.log NA NA
## S.npnct31.log S.npnct31.log NA NA
## S.npnct32.log S.npnct32.log NA NA
## S.nwrds.log S.nwrds.log NA 1.359149e-01
## S.nwrds.unq.log S.nwrds.unq.log NA -2.507969e-01
## S.T.articl S.T.articl NA -5.446201e-02
## S.T.compani S.T.compani NA -4.764341e-02
## S.T.fashion S.T.fashion NA -8.419711e-02
## S.T.first S.T.first NA -4.617532e-02
## S.T.intern S.T.intern NA -6.953750e-02
## S.T.newyork S.T.newyork NA -5.712853e-02
## S.T.report S.T.report NA -4.746920e-02
## S.T.week S.T.week NA -8.503373e-02
## S.T.will S.T.will NA -3.892267e-02
## S.T.year S.T.year NA -4.729011e-02
## UniqueID UniqueID NA 1.182492e-02
## WordCount WordCount NA 2.575265e-01
## exclude.as.feat cor.y.abs cor.high.X
## WordCount.log FALSE 2.656836e-01 <NA>
## myCategory.fctr FALSE 1.234541e-02 <NA>
## H.npnct21.log FALSE 1.283641e-01 <NA>
## A.npnct21.log FALSE 5.482747e-02 <NA>
## S.nuppr.log FALSE 2.718459e-01 <NA>
## H.nchrs.log FALSE 1.710624e-01 <NA>
## A.npnct14.log FALSE 4.999563e-02 <NA>
## H.npnct09.log FALSE 5.375262e-02 <NA>
## PubDate.wkday.fctr FALSE 3.980129e-02 <NA>
## H.nuppr.log FALSE 1.278085e-01 <NA>
## H.T.polit FALSE 3.062866e-02 <NA>
## A.T.newyork FALSE 5.706083e-02 <NA>
## A.T.report FALSE 4.741555e-02 <NA>
## H.npnct12.log FALSE 1.333613e-02 <NA>
## S.ndgts.log FALSE 1.242046e-01 <NA>
## A.nwrds.unq.log FALSE 2.506012e-01 <NA>
## S.T.said FALSE 1.826884e-02 <NA>
## S.T.make FALSE 3.959645e-02 <NA>
## H.npnct17.log FALSE 3.039622e-02 <NA>
## PubDate.last10.log FALSE 4.931702e-02 <NA>
## S.T.state FALSE 1.012205e-02 <NA>
## S.T.one FALSE 1.080534e-02 <NA>
## S.npnct04.log FALSE 6.294642e-02 <NA>
## S.T.share FALSE 5.070234e-02 <NA>
## PubDate.hour.fctr FALSE 1.354368e-01 <NA>
## H.T.report FALSE 6.244050e-02 <NA>
## PubDate.second.fctr FALSE 1.187946e-02 <NA>
## H.T.newyork FALSE 5.650839e-02 <NA>
## S.T.can FALSE 3.049697e-02 <NA>
## H.T.say FALSE 9.960773e-03 <NA>
## A.T.week FALSE 8.492895e-02 <NA>
## H.T.fashion FALSE 7.947505e-02 <NA>
## H.npnct16.log FALSE 8.273237e-02 <NA>
## A.T.will FALSE 3.887937e-02 <NA>
## S.T.show FALSE 4.193803e-02 <NA>
## PubDate.minute.fctr FALSE 3.407385e-02 <NA>
## H.ndgts.log FALSE 1.196633e-01 <NA>
## S.npnct12.log FALSE 9.158156e-02 <NA>
## H.npnct07.log FALSE 1.201741e-02 <NA>
## A.nwrds.log FALSE 1.354108e-01 <NA>
## H.T.art FALSE 3.280483e-02 <NA>
## H.npnct01.log FALSE 2.271577e-02 <NA>
## .rnorm FALSE 8.244230e-03 <NA>
## S.T.time FALSE 5.303654e-02 <NA>
## H.npnct13.log FALSE 1.305305e-02 <NA>
## S.npnct01.log FALSE 3.093101e-02 <NA>
## A.T.intern FALSE 6.949870e-02 <NA>
## A.T.compani FALSE 4.751471e-02 <NA>
## H.T.new FALSE 4.327803e-02 <NA>
## A.nchrs.log FALSE 2.245488e-01 <NA>
## PubDate.last1.log FALSE 4.635751e-02 <NA>
## H.npnct14.log FALSE 2.524770e-02 <NA>
## H.T.first FALSE 4.458885e-02 <NA>
## PubDate.date.fctr FALSE 1.164756e-02 <NA>
## H.T.busi FALSE 4.901905e-02 <NA>
## S.npnct13.log FALSE 3.638891e-02 <NA>
## H.T.day FALSE 6.033488e-02 <NA>
## H.T.china FALSE 3.144808e-02 <NA>
## A.T.year FALSE 4.721236e-02 <NA>
## H.nwrds.log FALSE 1.410282e-01 <NA>
## PubDate.wkend FALSE 1.067288e-01 <NA>
## H.T.big FALSE 1.390748e-02 <NA>
## H.T.news FALSE 4.415284e-02 <NA>
## H.T.week FALSE 6.812724e-02 <NA>
## H.has.ebola FALSE 2.588140e-02 <NA>
## H.T.take FALSE 8.582583e-04 <NA>
## H.T.bank FALSE 9.989139e-03 <NA>
## S.T.take FALSE 2.275732e-02 <NA>
## S.T.day FALSE 4.188671e-02 <NA>
## H.T.make FALSE 1.430572e-02 <NA>
## H.T.time FALSE 2.527450e-03 <NA>
## H.T.pictur FALSE 3.993172e-02 <NA>
## H.T.billion FALSE 2.949817e-02 <NA>
## H.T.X2014 FALSE 4.497745e-02 <NA>
## A.T.articl FALSE 5.445243e-02 <NA>
## H.T.obama FALSE 9.907543e-03 <NA>
## S.npnct16.log FALSE 6.770952e-02 <NA>
## PubDate.last100.log FALSE 7.663322e-03 <NA>
## S.npnct15.log FALSE 2.121844e-02 <NA>
## A.T.first FALSE 4.603341e-02 <NA>
## H.T.test FALSE 2.065255e-02 <NA>
## A.npnct17.log FALSE 1.587454e-03 <NA>
## S.npnct06.log FALSE 2.389145e-02 <NA>
## S.T.new FALSE 2.769558e-02 <NA>
## H.npnct30.log FALSE 8.917338e-02 <NA>
## A.T.fashion FALSE 8.419345e-02 <NA>
## H.T.springsumm FALSE 5.943248e-02 <NA>
## H.T.deal FALSE 2.559418e-02 <NA>
## S.npnct30.log FALSE 4.370037e-02 <NA>
## H.npnct02.log FALSE 2.001851e-02 <NA>
## S.npnct22.log FALSE 1.923169e-02 <NA>
## S.npnct07.log FALSE 1.214357e-02 <NA>
## A.T.presid FALSE 1.789086e-03 <NA>
## S.T.presid FALSE 2.079562e-03 <NA>
## S.npnct03.log FALSE 1.240734e-02 <NA>
## S.has.year.colon FALSE 1.755336e-02 <NA>
## H.npnct05.log FALSE 9.653967e-03 <NA>
## S.npnct08.log FALSE 2.413868e-03 <NA>
## S.npnct09.log FALSE 3.986882e-03 <NA>
## A.npnct19.log FALSE 1.271661e-02 <NA>
## A.npnct20.log FALSE 1.451467e-02 <NA>
## A.has.http FALSE 1.359260e-02 A.npnct19.log
## A.has.year.colon FALSE 1.755336e-02 S.has.year.colon
## A.ndgts.log FALSE 1.249484e-01 S.ndgts.log
## A.npnct01.log FALSE 3.093101e-02 S.npnct01.log
## A.npnct02.log FALSE 1.451467e-02 A.npnct18.log
## A.npnct03.log FALSE 1.359260e-02 S.npnct03.log
## A.npnct04.log FALSE 6.294642e-02 S.npnct04.log
## A.npnct05.log FALSE NA <NA>
## A.npnct06.log FALSE 2.389145e-02 S.npnct06.log
## A.npnct07.log FALSE 1.214357e-02 S.npnct07.log
## A.npnct08.log FALSE 3.258100e-03 <NA>
## A.npnct09.log FALSE 4.775988e-03 <NA>
## A.npnct10.log FALSE NA <NA>
## A.npnct11.log FALSE 5.547032e-03 <NA>
## A.npnct12.log FALSE 9.183870e-02 S.npnct12.log
## A.npnct13.log FALSE 3.760012e-02 S.npnct13.log
## A.npnct15.log FALSE 2.407715e-02 A.npnct02.log
## A.npnct16.log FALSE 6.893301e-02 S.npnct16.log
## A.npnct18.log FALSE 1.451467e-02 A.npnct20.log
## A.npnct22.log FALSE 1.923169e-02 S.npnct22.log
## A.npnct23.log FALSE 1.537569e-02 A.npnct25.log
## A.npnct24.log FALSE NA <NA>
## A.npnct25.log FALSE 1.537569e-02 <NA>
## A.npnct26.log FALSE 9.890046e-19 <NA>
## A.npnct27.log FALSE 5.547032e-03 <NA>
## A.npnct28.log FALSE NA <NA>
## A.npnct29.log FALSE NA <NA>
## A.npnct30.log FALSE 4.373349e-02 S.npnct30.log
## A.npnct31.log FALSE NA <NA>
## A.npnct32.log FALSE NA <NA>
## A.nuppr.log FALSE 2.720962e-01 S.nuppr.log
## A.T.can FALSE 3.127063e-02 S.T.can
## A.T.day FALSE 4.196599e-02 S.T.day
## A.T.make FALSE 3.965722e-02 S.T.make
## A.T.new FALSE 2.782876e-02 S.T.new
## A.T.one FALSE 1.081694e-02 S.T.one
## A.T.said FALSE 1.839871e-02 S.T.said
## A.T.share FALSE 5.070234e-02 S.T.share
## A.T.show FALSE 4.196129e-02 S.T.show
## A.T.state FALSE 1.020706e-02 S.T.state
## A.T.take FALSE 2.282555e-02 S.T.take
## A.T.time FALSE 5.313395e-02 S.T.time
## clusterid FALSE NA <NA>
## H.has.http FALSE NA <NA>
## H.has.year.colon FALSE 7.842875e-02 S.T.intern
## H.npnct03.log FALSE 9.533020e-03 <NA>
## H.npnct04.log FALSE 5.126277e-02 H.T.billion
## H.npnct06.log FALSE 3.190718e-02 H.npnct17.log
## H.npnct08.log FALSE 5.375262e-02 H.npnct09.log
## H.npnct10.log FALSE NA <NA>
## H.npnct11.log FALSE 5.547032e-03 <NA>
## H.npnct15.log FALSE 6.158577e-02 H.T.springsumm
## H.npnct18.log FALSE NA <NA>
## H.npnct19.log FALSE NA <NA>
## H.npnct20.log FALSE NA <NA>
## H.npnct22.log FALSE 5.547032e-03 <NA>
## H.npnct23.log FALSE NA <NA>
## H.npnct24.log FALSE NA <NA>
## H.npnct25.log FALSE NA <NA>
## H.npnct26.log FALSE 9.890046e-19 <NA>
## H.npnct27.log FALSE NA <NA>
## H.npnct28.log FALSE NA <NA>
## H.npnct29.log FALSE NA <NA>
## H.npnct31.log FALSE NA <NA>
## H.npnct32.log FALSE NA <NA>
## H.nwrds.unq.log FALSE 2.044964e-01 H.nuppr.log
## H.T.daili FALSE 6.299948e-02 H.T.report
## H.T.morn FALSE 4.838380e-02 A.npnct30.log
## H.T.today FALSE 5.831308e-02 H.T.polit
## H.T.X2015 FALSE 6.601141e-02 H.npnct15.log
## Popular TRUE 1.000000e+00 <NA>
## Popular.fctr TRUE NA <NA>
## PubDate.last1 TRUE 3.592267e-02 <NA>
## PubDate.last10 TRUE 5.398093e-02 <NA>
## PubDate.last100 TRUE 3.989229e-02 <NA>
## PubDate.month.fctr TRUE 1.914874e-02 <NA>
## PubDate.POSIX TRUE 1.568326e-02 <NA>
## PubDate.year.fctr FALSE NA <NA>
## PubDate.zoo TRUE 1.568326e-02 <NA>
## S.has.http FALSE NA <NA>
## S.nchrs.log FALSE 2.246930e-01 A.nchrs.log
## S.npnct02.log FALSE 5.547032e-03 <NA>
## S.npnct05.log FALSE NA <NA>
## S.npnct10.log FALSE NA <NA>
## S.npnct11.log FALSE 5.547032e-03 <NA>
## S.npnct14.log FALSE 5.332519e-02 A.npnct14.log
## S.npnct17.log FALSE 1.587454e-03 <NA>
## S.npnct18.log FALSE NA <NA>
## S.npnct19.log FALSE NA <NA>
## S.npnct20.log FALSE NA <NA>
## S.npnct21.log FALSE 5.503894e-02 A.npnct21.log
## S.npnct23.log FALSE 2.760321e-02 A.npnct23.log
## S.npnct24.log FALSE NA <NA>
## S.npnct25.log FALSE 2.760321e-02 <NA>
## S.npnct26.log FALSE 9.890046e-19 <NA>
## S.npnct27.log FALSE NA <NA>
## S.npnct28.log FALSE NA <NA>
## S.npnct29.log FALSE NA <NA>
## S.npnct31.log FALSE NA <NA>
## S.npnct32.log FALSE NA <NA>
## S.nwrds.log FALSE 1.359149e-01 A.nwrds.log
## S.nwrds.unq.log FALSE 2.507969e-01 S.nchrs.log
## S.T.articl FALSE 5.446201e-02 A.T.articl
## S.T.compani FALSE 4.764341e-02 A.T.compani
## S.T.fashion FALSE 8.419711e-02 H.T.X2015
## S.T.first FALSE 4.617532e-02 A.T.first
## S.T.intern FALSE 6.953750e-02 A.T.intern
## S.T.newyork FALSE 5.712853e-02 A.T.newyork
## S.T.report FALSE 4.746920e-02 A.T.report
## S.T.week FALSE 8.503373e-02 A.T.week
## S.T.will FALSE 3.892267e-02 A.T.will
## S.T.year FALSE 4.729011e-02 A.T.year
## UniqueID TRUE 1.182492e-02 <NA>
## WordCount TRUE 2.575265e-01 <NA>
## freqRatio percentUnique zeroVar nzv myNearZV
## WordCount.log 1.300000 24.14268218 FALSE FALSE FALSE
## myCategory.fctr 1.337185 0.30618494 FALSE FALSE FALSE
## H.npnct21.log 14.995098 0.06123699 FALSE FALSE FALSE
## A.npnct21.log 12.798715 0.07654623 FALSE FALSE FALSE
## S.nuppr.log 1.152620 0.33680343 FALSE FALSE FALSE
## H.nchrs.log 1.023810 1.57685242 FALSE FALSE FALSE
## A.npnct14.log 4.603330 0.16840171 FALSE FALSE FALSE
## H.npnct09.log 111.620690 0.03061849 FALSE TRUE FALSE
## PubDate.wkday.fctr 1.003268 0.10716473 FALSE FALSE FALSE
## H.nuppr.log 1.033930 0.29087569 FALSE FALSE FALSE
## H.T.polit 126.254902 0.13778322 FALSE TRUE FALSE
## A.T.newyork 88.724638 0.42865891 FALSE TRUE FALSE
## A.T.report 78.362500 0.38273117 FALSE TRUE FALSE
## H.npnct12.log 4.937442 0.07654623 FALSE FALSE FALSE
## S.ndgts.log 10.511247 0.26025720 FALSE FALSE FALSE
## A.nwrds.unq.log 1.061567 0.55113288 FALSE FALSE FALSE
## S.T.said 190.242424 0.36742192 FALSE TRUE FALSE
## S.T.make 273.782609 0.47458665 FALSE TRUE FALSE
## H.npnct17.log 96.104478 0.06123699 FALSE TRUE FALSE
## PubDate.last10.log 1.666667 79.05695040 FALSE FALSE FALSE
## S.T.state 315.750000 0.42865891 FALSE TRUE FALSE
## S.T.one 240.038462 0.45927740 FALSE TRUE FALSE
## S.npnct04.log 28.536364 0.07654623 FALSE TRUE FALSE
## S.T.share 218.448276 0.36742192 FALSE TRUE FALSE
## PubDate.hour.fctr 1.835040 0.04592774 FALSE FALSE FALSE
## H.T.report 102.000000 0.16840171 FALSE TRUE FALSE
## PubDate.second.fctr 1.018204 0.06123699 FALSE FALSE FALSE
## H.T.newyork 112.446429 0.15309247 FALSE TRUE FALSE
## S.T.can 261.666667 0.41334966 FALSE TRUE FALSE
## H.T.say 247.461538 0.16840171 FALSE TRUE FALSE
## A.T.week 57.122642 0.48989590 FALSE TRUE FALSE
## H.T.fashion 76.926829 0.19902021 FALSE TRUE FALSE
## H.npnct16.log 3.914910 0.04592774 FALSE FALSE FALSE
## A.T.will 114.711538 0.62767912 FALSE TRUE FALSE
## S.T.show 274.608696 0.39804042 FALSE TRUE FALSE
## PubDate.minute.fctr 1.483365 0.06123699 FALSE FALSE FALSE
## H.ndgts.log 13.616137 0.18371096 FALSE FALSE FALSE
## S.npnct12.log 1.660473 0.13778322 FALSE FALSE FALSE
## H.npnct07.log 5.437234 0.12247397 FALSE FALSE FALSE
## A.nwrds.log 2.583333 93.35578690 FALSE FALSE FALSE
## H.T.art 307.333333 0.19902021 FALSE TRUE FALSE
## H.npnct01.log 282.913043 0.04592774 FALSE TRUE FALSE
## .rnorm 2.000000 99.98469075 FALSE FALSE FALSE
## S.T.time 65.096774 0.47458665 FALSE TRUE FALSE
## H.npnct13.log 13.126638 0.09185548 FALSE FALSE FALSE
## S.npnct01.log 309.952381 0.06123699 FALSE TRUE FALSE
## A.T.intern 137.347826 0.32149418 FALSE TRUE FALSE
## A.T.compani 140.227273 0.50520514 FALSE TRUE FALSE
## H.T.new 116.333333 0.19902021 FALSE TRUE FALSE
## A.nchrs.log 1.328571 4.39375383 FALSE FALSE FALSE
## PubDate.last1.log 1.142857 36.49724434 FALSE FALSE FALSE
## H.npnct14.log 22.802326 0.12247397 FALSE TRUE FALSE
## H.T.first 194.727273 0.15309247 FALSE TRUE FALSE
## PubDate.date.fctr 1.021394 0.07654623 FALSE FALSE FALSE
## H.T.busi 229.428571 0.18371096 FALSE TRUE FALSE
## S.npnct13.log 5.706263 0.09185548 FALSE FALSE FALSE
## H.T.day 86.547945 0.18371096 FALSE TRUE FALSE
## H.T.china 238.555556 0.18371096 FALSE TRUE FALSE
## A.T.year 167.108108 0.44396816 FALSE TRUE FALSE
## H.nwrds.log 1.127273 84.12431108 FALSE FALSE FALSE
## PubDate.wkend 9.095827 0.03061849 FALSE FALSE FALSE
## H.T.big 403.562500 0.19902021 FALSE TRUE FALSE
## H.T.news 238.518519 0.16840171 FALSE TRUE FALSE
## H.T.week 64.071429 0.16840171 FALSE TRUE FALSE
## H.has.ebola 73.227273 0.03061849 FALSE TRUE FALSE
## H.T.take 306.904762 0.15309247 FALSE TRUE FALSE
## H.T.bank 221.689655 0.13778322 FALSE TRUE FALSE
## S.T.take 287.090909 0.38273117 FALSE TRUE FALSE
## S.T.day 89.600000 0.39804042 FALSE TRUE FALSE
## H.T.make 322.200000 0.13778322 FALSE TRUE FALSE
## H.T.time 247.538462 0.16840171 FALSE TRUE FALSE
## H.T.pictur 104.032258 0.10716473 FALSE TRUE FALSE
## H.T.billion 229.892857 0.13778322 FALSE TRUE FALSE
## H.T.X2014 112.824561 0.13778322 FALSE TRUE FALSE
## A.T.articl 85.500000 0.27556644 FALSE TRUE FALSE
## H.T.obama 229.750000 0.16840171 FALSE TRUE FALSE
## S.npnct16.log 13.647191 0.04592774 FALSE FALSE FALSE
## PubDate.last100.log 25.000000 92.19228414 FALSE FALSE FALSE
## S.npnct15.log 203.062500 0.04592774 FALSE TRUE FALSE
## A.T.first 203.709677 0.42865891 FALSE TRUE FALSE
## H.T.test 280.000000 0.13778322 FALSE TRUE FALSE
## A.npnct17.log 434.133333 0.04592774 FALSE TRUE FALSE
## S.npnct06.log 115.642857 0.03061849 FALSE TRUE FALSE
## S.T.new 107.872727 0.48989590 FALSE TRUE FALSE
## H.npnct30.log 24.123077 0.03061849 FALSE TRUE FALSE
## A.T.fashion 59.809524 0.41334966 FALSE TRUE FALSE
## H.T.springsumm 106.966667 0.09185548 FALSE TRUE FALSE
## H.T.deal 258.080000 0.13778322 FALSE TRUE FALSE
## S.npnct30.log 134.791667 0.04592774 FALSE TRUE FALSE
## H.npnct02.log 501.461538 0.03061849 FALSE TRUE FALSE
## S.npnct22.log 543.333333 0.03061849 FALSE TRUE FALSE
## S.npnct07.log 1631.750000 0.04592774 FALSE TRUE FALSE
## A.T.presid 241.692308 0.45927740 FALSE TRUE FALSE
## S.T.presid 241.692308 0.42865891 FALSE TRUE FALSE
## S.npnct03.log 1305.400000 0.03061849 FALSE TRUE FALSE
## S.has.year.colon 652.200000 0.03061849 FALSE TRUE FALSE
## H.npnct05.log 543.333333 0.03061849 FALSE TRUE FALSE
## S.npnct08.log 175.513514 0.04592774 FALSE TRUE FALSE
## S.npnct09.log 175.486486 0.06123699 FALSE TRUE FALSE
## A.npnct19.log 1631.500000 0.06123699 FALSE TRUE FALSE
## A.npnct20.log 1087.500000 0.04592774 FALSE TRUE FALSE
## A.has.http 1087.666667 0.03061849 FALSE TRUE FALSE
## A.has.year.colon 652.200000 0.03061849 FALSE TRUE FALSE
## A.ndgts.log 10.501022 0.29087569 FALSE FALSE FALSE
## A.npnct01.log 309.952381 0.06123699 FALSE TRUE FALSE
## A.npnct02.log 1087.500000 0.04592774 FALSE TRUE FALSE
## A.npnct03.log 1087.666667 0.03061849 FALSE TRUE FALSE
## A.npnct04.log 28.536364 0.07654623 FALSE TRUE FALSE
## A.npnct05.log 0.000000 0.01530925 TRUE TRUE TRUE
## A.npnct06.log 115.642857 0.03061849 FALSE TRUE FALSE
## A.npnct07.log 1631.750000 0.04592774 FALSE TRUE FALSE
## A.npnct08.log 170.868421 0.04592774 FALSE TRUE FALSE
## A.npnct09.log 170.842105 0.06123699 FALSE TRUE FALSE
## A.npnct10.log 0.000000 0.01530925 TRUE TRUE TRUE
## A.npnct11.log 6531.000000 0.03061849 FALSE TRUE TRUE
## A.npnct12.log 1.660473 0.13778322 FALSE FALSE FALSE
## A.npnct13.log 5.715368 0.12247397 FALSE FALSE FALSE
## A.npnct15.log 196.696970 0.10716473 FALSE TRUE FALSE
## A.npnct16.log 13.482222 0.04592774 FALSE FALSE FALSE
## A.npnct18.log 1087.500000 0.04592774 FALSE TRUE FALSE
## A.npnct22.log 543.333333 0.03061849 FALSE TRUE FALSE
## A.npnct23.log 3264.500000 0.04592774 FALSE TRUE TRUE
## A.npnct24.log 0.000000 0.01530925 TRUE TRUE TRUE
## A.npnct25.log 3264.500000 0.04592774 FALSE TRUE TRUE
## A.npnct26.log 0.000000 0.01530925 TRUE TRUE TRUE
## A.npnct27.log 6531.000000 0.03061849 FALSE TRUE TRUE
## A.npnct28.log 0.000000 0.01530925 TRUE TRUE TRUE
## A.npnct29.log 0.000000 0.01530925 TRUE TRUE TRUE
## A.npnct30.log 126.862745 0.04592774 FALSE TRUE FALSE
## A.npnct31.log 0.000000 0.01530925 TRUE TRUE TRUE
## A.npnct32.log 0.000000 0.01530925 TRUE TRUE TRUE
## A.nuppr.log 1.151308 0.33680343 FALSE FALSE FALSE
## A.T.can 261.666667 0.48989590 FALSE TRUE FALSE
## A.T.day 89.585714 0.42865891 FALSE TRUE FALSE
## A.T.make 273.782609 0.48989590 FALSE TRUE FALSE
## A.T.new 107.836364 0.50520514 FALSE TRUE FALSE
## A.T.one 240.000000 0.50520514 FALSE TRUE FALSE
## A.T.said 190.242424 0.39804042 FALSE TRUE FALSE
## A.T.share 218.448276 0.36742192 FALSE TRUE FALSE
## A.T.show 274.608696 0.41334966 FALSE TRUE FALSE
## A.T.state 315.700000 0.42865891 FALSE TRUE FALSE
## A.T.take 287.045455 0.42865891 FALSE TRUE FALSE
## A.T.time 65.086022 0.47458665 FALSE TRUE FALSE
## clusterid 0.000000 0.01530925 TRUE TRUE TRUE
## H.has.http 0.000000 0.01530925 TRUE TRUE TRUE
## H.has.year.colon 32.670103 0.03061849 FALSE TRUE FALSE
## H.npnct03.log 2176.333333 0.03061849 FALSE TRUE TRUE
## H.npnct04.log 38.325301 0.04592774 FALSE TRUE FALSE
## H.npnct06.log 68.935484 0.06123699 FALSE TRUE FALSE
## H.npnct08.log 111.620690 0.03061849 FALSE TRUE FALSE
## H.npnct10.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct11.log 6531.000000 0.03061849 FALSE TRUE TRUE
## H.npnct15.log 52.983471 0.03061849 FALSE TRUE FALSE
## H.npnct18.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct19.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct20.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct22.log 6531.000000 0.03061849 FALSE TRUE TRUE
## H.npnct23.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct24.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct25.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct26.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct27.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct28.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct29.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct31.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.npnct32.log 0.000000 0.01530925 TRUE TRUE TRUE
## H.nwrds.unq.log 1.019071 0.21432945 FALSE FALSE FALSE
## H.T.daili 102.903226 0.16840171 FALSE TRUE FALSE
## H.T.morn 165.205128 0.07654623 FALSE TRUE FALSE
## H.T.today 138.239130 0.13778322 FALSE TRUE FALSE
## H.T.X2015 96.833333 0.10716473 FALSE TRUE FALSE
## Popular 4.976212 0.03061849 FALSE FALSE FALSE
## Popular.fctr NA NA NA NA NA
## PubDate.last1 1.142857 36.49724434 FALSE FALSE FALSE
## PubDate.last10 1.666667 79.05695040 FALSE FALSE FALSE
## PubDate.last100 25.000000 92.52908757 FALSE FALSE FALSE
## PubDate.month.fctr 1.017514 0.04592774 FALSE FALSE FALSE
## PubDate.POSIX 1.000000 99.86221678 FALSE FALSE FALSE
## PubDate.year.fctr 0.000000 0.01530925 TRUE TRUE TRUE
## PubDate.zoo 1.000000 99.86221678 FALSE FALSE FALSE
## S.has.http 0.000000 0.01530925 TRUE TRUE TRUE
## S.nchrs.log 1.328571 3.72014697 FALSE FALSE FALSE
## S.npnct02.log 6531.000000 0.03061849 FALSE TRUE TRUE
## S.npnct05.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct10.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct11.log 6531.000000 0.03061849 FALSE TRUE TRUE
## S.npnct14.log 4.672000 0.16840171 FALSE FALSE FALSE
## S.npnct17.log 434.133333 0.04592774 FALSE TRUE FALSE
## S.npnct18.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct19.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct20.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct21.log 12.862366 0.07654623 FALSE FALSE FALSE
## S.npnct23.log 6531.000000 0.03061849 FALSE TRUE TRUE
## S.npnct24.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct25.log 6531.000000 0.03061849 FALSE TRUE TRUE
## S.npnct26.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct27.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct28.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct29.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct31.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.npnct32.log 0.000000 0.01530925 TRUE TRUE TRUE
## S.nwrds.log 2.583333 93.67728108 FALSE FALSE FALSE
## S.nwrds.unq.log 1.061567 0.44396816 FALSE FALSE FALSE
## S.T.articl 85.500000 0.27556644 FALSE TRUE FALSE
## S.T.compani 140.227273 0.44396816 FALSE TRUE FALSE
## S.T.fashion 59.809524 0.39804042 FALSE TRUE FALSE
## S.T.first 203.709677 0.39804042 FALSE TRUE FALSE
## S.T.intern 137.347826 0.30618494 FALSE TRUE FALSE
## S.T.newyork 88.724638 0.41334966 FALSE TRUE FALSE
## S.T.report 78.362500 0.35211268 FALSE TRUE FALSE
## S.T.week 57.122642 0.42865891 FALSE TRUE FALSE
## S.T.will 114.750000 0.55113288 FALSE TRUE FALSE
## S.T.year 167.108108 0.42865891 FALSE TRUE FALSE
## UniqueID 1.000000 100.00000000 FALSE FALSE FALSE
## WordCount 2.315789 24.15799143 FALSE FALSE FALSE
## is.cor.y.abs.low rsp_var_raw id_var rsp_var
## WordCount.log FALSE FALSE NA NA
## myCategory.fctr FALSE FALSE NA NA
## H.npnct21.log FALSE FALSE NA NA
## A.npnct21.log FALSE FALSE NA NA
## S.nuppr.log FALSE FALSE NA NA
## H.nchrs.log FALSE FALSE NA NA
## A.npnct14.log FALSE FALSE NA NA
## H.npnct09.log FALSE FALSE NA NA
## PubDate.wkday.fctr FALSE FALSE NA NA
## H.nuppr.log FALSE FALSE NA NA
## H.T.polit FALSE FALSE NA NA
## A.T.newyork FALSE FALSE NA NA
## A.T.report FALSE FALSE NA NA
## H.npnct12.log FALSE FALSE NA NA
## S.ndgts.log FALSE FALSE NA NA
## A.nwrds.unq.log FALSE FALSE NA NA
## S.T.said FALSE FALSE NA NA
## S.T.make FALSE FALSE NA NA
## H.npnct17.log FALSE FALSE NA NA
## PubDate.last10.log FALSE FALSE NA NA
## S.T.state FALSE FALSE NA NA
## S.T.one FALSE FALSE NA NA
## S.npnct04.log FALSE FALSE NA NA
## S.T.share FALSE FALSE NA NA
## PubDate.hour.fctr FALSE FALSE NA NA
## H.T.report FALSE FALSE NA NA
## PubDate.second.fctr FALSE FALSE NA NA
## H.T.newyork FALSE FALSE NA NA
## S.T.can FALSE FALSE NA NA
## H.T.say FALSE FALSE NA NA
## A.T.week FALSE FALSE NA NA
## H.T.fashion FALSE FALSE NA NA
## H.npnct16.log FALSE FALSE NA NA
## A.T.will FALSE FALSE NA NA
## S.T.show FALSE FALSE NA NA
## PubDate.minute.fctr FALSE FALSE NA NA
## H.ndgts.log FALSE FALSE NA NA
## S.npnct12.log FALSE FALSE NA NA
## H.npnct07.log FALSE FALSE NA NA
## A.nwrds.log FALSE FALSE NA NA
## H.T.art FALSE FALSE NA NA
## H.npnct01.log FALSE FALSE NA NA
## .rnorm FALSE FALSE NA NA
## S.T.time FALSE FALSE NA NA
## H.npnct13.log FALSE FALSE NA NA
## S.npnct01.log FALSE FALSE NA NA
## A.T.intern FALSE FALSE NA NA
## A.T.compani FALSE FALSE NA NA
## H.T.new FALSE FALSE NA NA
## A.nchrs.log FALSE FALSE NA NA
## PubDate.last1.log FALSE FALSE NA NA
## H.npnct14.log FALSE FALSE NA NA
## H.T.first FALSE FALSE NA NA
## PubDate.date.fctr FALSE FALSE NA NA
## H.T.busi FALSE FALSE NA NA
## S.npnct13.log FALSE FALSE NA NA
## H.T.day FALSE FALSE NA NA
## H.T.china FALSE FALSE NA NA
## A.T.year FALSE FALSE NA NA
## H.nwrds.log FALSE FALSE NA NA
## PubDate.wkend FALSE FALSE NA NA
## H.T.big FALSE FALSE NA NA
## H.T.news FALSE FALSE NA NA
## H.T.week FALSE FALSE NA NA
## H.has.ebola FALSE FALSE NA NA
## H.T.take TRUE FALSE NA NA
## H.T.bank FALSE FALSE NA NA
## S.T.take FALSE FALSE NA NA
## S.T.day FALSE FALSE NA NA
## H.T.make FALSE FALSE NA NA
## H.T.time TRUE FALSE NA NA
## H.T.pictur FALSE FALSE NA NA
## H.T.billion FALSE FALSE NA NA
## H.T.X2014 FALSE FALSE NA NA
## A.T.articl FALSE FALSE NA NA
## H.T.obama FALSE FALSE NA NA
## S.npnct16.log FALSE FALSE NA NA
## PubDate.last100.log TRUE FALSE NA NA
## S.npnct15.log FALSE FALSE NA NA
## A.T.first FALSE FALSE NA NA
## H.T.test FALSE FALSE NA NA
## A.npnct17.log TRUE FALSE NA NA
## S.npnct06.log FALSE FALSE NA NA
## S.T.new FALSE FALSE NA NA
## H.npnct30.log FALSE FALSE NA NA
## A.T.fashion FALSE FALSE NA NA
## H.T.springsumm FALSE FALSE NA NA
## H.T.deal FALSE FALSE NA NA
## S.npnct30.log FALSE FALSE NA NA
## H.npnct02.log FALSE FALSE NA NA
## S.npnct22.log FALSE FALSE NA NA
## S.npnct07.log FALSE FALSE NA NA
## A.T.presid TRUE FALSE NA NA
## S.T.presid TRUE FALSE NA NA
## S.npnct03.log FALSE FALSE NA NA
## S.has.year.colon FALSE FALSE NA NA
## H.npnct05.log FALSE FALSE NA NA
## S.npnct08.log TRUE FALSE NA NA
## S.npnct09.log TRUE FALSE NA NA
## A.npnct19.log FALSE FALSE NA NA
## A.npnct20.log FALSE FALSE NA NA
## A.has.http FALSE FALSE NA NA
## A.has.year.colon FALSE FALSE NA NA
## A.ndgts.log FALSE FALSE NA NA
## A.npnct01.log FALSE FALSE NA NA
## A.npnct02.log FALSE FALSE NA NA
## A.npnct03.log FALSE FALSE NA NA
## A.npnct04.log FALSE FALSE NA NA
## A.npnct05.log NA FALSE NA NA
## A.npnct06.log FALSE FALSE NA NA
## A.npnct07.log FALSE FALSE NA NA
## A.npnct08.log TRUE FALSE NA NA
## A.npnct09.log TRUE FALSE NA NA
## A.npnct10.log NA FALSE NA NA
## A.npnct11.log TRUE FALSE NA NA
## A.npnct12.log FALSE FALSE NA NA
## A.npnct13.log FALSE FALSE NA NA
## A.npnct15.log FALSE FALSE NA NA
## A.npnct16.log FALSE FALSE NA NA
## A.npnct18.log FALSE FALSE NA NA
## A.npnct22.log FALSE FALSE NA NA
## A.npnct23.log FALSE FALSE NA NA
## A.npnct24.log NA FALSE NA NA
## A.npnct25.log FALSE FALSE NA NA
## A.npnct26.log TRUE FALSE NA NA
## A.npnct27.log TRUE FALSE NA NA
## A.npnct28.log NA FALSE NA NA
## A.npnct29.log NA FALSE NA NA
## A.npnct30.log FALSE FALSE NA NA
## A.npnct31.log NA FALSE NA NA
## A.npnct32.log NA FALSE NA NA
## A.nuppr.log FALSE FALSE NA NA
## A.T.can FALSE FALSE NA NA
## A.T.day FALSE FALSE NA NA
## A.T.make FALSE FALSE NA NA
## A.T.new FALSE FALSE NA NA
## A.T.one FALSE FALSE NA NA
## A.T.said FALSE FALSE NA NA
## A.T.share FALSE FALSE NA NA
## A.T.show FALSE FALSE NA NA
## A.T.state FALSE FALSE NA NA
## A.T.take FALSE FALSE NA NA
## A.T.time FALSE FALSE NA NA
## clusterid NA FALSE NA NA
## H.has.http NA FALSE NA NA
## H.has.year.colon FALSE FALSE NA NA
## H.npnct03.log FALSE FALSE NA NA
## H.npnct04.log FALSE FALSE NA NA
## H.npnct06.log FALSE FALSE NA NA
## H.npnct08.log FALSE FALSE NA NA
## H.npnct10.log NA FALSE NA NA
## H.npnct11.log TRUE FALSE NA NA
## H.npnct15.log FALSE FALSE NA NA
## H.npnct18.log NA FALSE NA NA
## H.npnct19.log NA FALSE NA NA
## H.npnct20.log NA FALSE NA NA
## H.npnct22.log TRUE FALSE NA NA
## H.npnct23.log NA FALSE NA NA
## H.npnct24.log NA FALSE NA NA
## H.npnct25.log NA FALSE NA NA
## H.npnct26.log TRUE FALSE NA NA
## H.npnct27.log NA FALSE NA NA
## H.npnct28.log NA FALSE NA NA
## H.npnct29.log NA FALSE NA NA
## H.npnct31.log NA FALSE NA NA
## H.npnct32.log NA FALSE NA NA
## H.nwrds.unq.log FALSE FALSE NA NA
## H.T.daili FALSE FALSE NA NA
## H.T.morn FALSE FALSE NA NA
## H.T.today FALSE FALSE NA NA
## H.T.X2015 FALSE FALSE NA NA
## Popular FALSE TRUE NA NA
## Popular.fctr NA NA NA TRUE
## PubDate.last1 FALSE FALSE NA NA
## PubDate.last10 FALSE FALSE NA NA
## PubDate.last100 FALSE FALSE NA NA
## PubDate.month.fctr FALSE FALSE NA NA
## PubDate.POSIX FALSE FALSE NA NA
## PubDate.year.fctr NA FALSE NA NA
## PubDate.zoo FALSE FALSE NA NA
## S.has.http NA FALSE NA NA
## S.nchrs.log FALSE FALSE NA NA
## S.npnct02.log TRUE FALSE NA NA
## S.npnct05.log NA FALSE NA NA
## S.npnct10.log NA FALSE NA NA
## S.npnct11.log TRUE FALSE NA NA
## S.npnct14.log FALSE FALSE NA NA
## S.npnct17.log TRUE FALSE NA NA
## S.npnct18.log NA FALSE NA NA
## S.npnct19.log NA FALSE NA NA
## S.npnct20.log NA FALSE NA NA
## S.npnct21.log FALSE FALSE NA NA
## S.npnct23.log FALSE FALSE NA NA
## S.npnct24.log NA FALSE NA NA
## S.npnct25.log FALSE FALSE NA NA
## S.npnct26.log TRUE FALSE NA NA
## S.npnct27.log NA FALSE NA NA
## S.npnct28.log NA FALSE NA NA
## S.npnct29.log NA FALSE NA NA
## S.npnct31.log NA FALSE NA NA
## S.npnct32.log NA FALSE NA NA
## S.nwrds.log FALSE FALSE NA NA
## S.nwrds.unq.log FALSE FALSE NA NA
## S.T.articl FALSE FALSE NA NA
## S.T.compani FALSE FALSE NA NA
## S.T.fashion FALSE FALSE NA NA
## S.T.first FALSE FALSE NA NA
## S.T.intern FALSE FALSE NA NA
## S.T.newyork FALSE FALSE NA NA
## S.T.report FALSE FALSE NA NA
## S.T.week FALSE FALSE NA NA
## S.T.will FALSE FALSE NA NA
## S.T.year FALSE FALSE NA NA
## UniqueID FALSE FALSE TRUE NA
## WordCount FALSE FALSE NA NA
## Low.cor.X.glm.importance Final.glm.importance
## WordCount.log 1.000000e+02 1.000000e+02
## myCategory.fctr 7.312143e+01 7.312143e+01
## H.npnct21.log 3.777302e+01 3.777302e+01
## A.npnct21.log 3.475205e+01 3.475205e+01
## S.nuppr.log 3.238555e+01 3.238555e+01
## H.nchrs.log 3.201395e+01 3.201395e+01
## A.npnct14.log 2.764418e+01 2.764418e+01
## H.npnct09.log 2.188969e+01 2.188969e+01
## PubDate.wkday.fctr 2.165613e+01 2.165613e+01
## H.nuppr.log 2.131190e+01 2.131190e+01
## H.T.polit 2.052535e+01 2.052535e+01
## A.T.newyork 1.961261e+01 1.961261e+01
## A.T.report 1.938341e+01 1.938341e+01
## H.npnct12.log 1.796841e+01 1.796841e+01
## S.ndgts.log 1.621363e+01 1.621363e+01
## A.nwrds.unq.log 1.619197e+01 1.619197e+01
## S.T.said 1.520654e+01 1.520654e+01
## S.T.make 1.512129e+01 1.512129e+01
## H.npnct17.log 1.480733e+01 1.480733e+01
## PubDate.last10.log 1.453274e+01 1.453274e+01
## S.T.state 1.414282e+01 1.414282e+01
## S.T.one 1.375899e+01 1.375899e+01
## S.npnct04.log 1.373898e+01 1.373898e+01
## S.T.share 1.314940e+01 1.314940e+01
## PubDate.hour.fctr 1.314384e+01 1.314384e+01
## H.T.report 1.299102e+01 1.299102e+01
## PubDate.second.fctr 1.273156e+01 1.273156e+01
## H.T.newyork 1.269274e+01 1.269274e+01
## S.T.can 1.228026e+01 1.228026e+01
## H.T.say 1.222772e+01 1.222772e+01
## A.T.week 1.168514e+01 1.168514e+01
## H.T.fashion 1.151035e+01 1.151035e+01
## H.npnct16.log 1.136051e+01 1.136051e+01
## A.T.will 1.129252e+01 1.129252e+01
## S.T.show 1.089884e+01 1.089884e+01
## PubDate.minute.fctr 1.041575e+01 1.041575e+01
## H.ndgts.log 1.035074e+01 1.035074e+01
## S.npnct12.log 9.930532e+00 9.930532e+00
## H.npnct07.log 9.854654e+00 9.854654e+00
## A.nwrds.log 9.770236e+00 9.770236e+00
## H.T.art 9.623245e+00 9.623245e+00
## H.npnct01.log 9.509217e+00 9.509217e+00
## .rnorm 9.238260e+00 9.238260e+00
## S.T.time 9.237301e+00 9.237301e+00
## H.npnct13.log 9.109144e+00 9.109144e+00
## S.npnct01.log 8.749827e+00 8.749827e+00
## A.T.intern 8.472657e+00 8.472657e+00
## A.T.compani 8.369682e+00 8.369682e+00
## H.T.new 8.270503e+00 8.270503e+00
## A.nchrs.log 8.248780e+00 8.248780e+00
## PubDate.last1.log 8.146784e+00 8.146784e+00
## H.npnct14.log 8.031911e+00 8.031911e+00
## H.T.first 7.798393e+00 7.798393e+00
## PubDate.date.fctr 7.300560e+00 7.300560e+00
## H.T.busi 7.168757e+00 7.168757e+00
## S.npnct13.log 6.710517e+00 6.710517e+00
## H.T.day 6.378120e+00 6.378120e+00
## H.T.china 6.350450e+00 6.350450e+00
## A.T.year 6.178982e+00 6.178982e+00
## H.nwrds.log 6.165115e+00 6.165115e+00
## PubDate.wkend 6.020520e+00 6.020520e+00
## H.T.big 6.007063e+00 6.007063e+00
## H.T.news 5.524332e+00 5.524332e+00
## H.T.week 5.347019e+00 5.347019e+00
## H.has.ebola 4.853552e+00 4.853552e+00
## H.T.take 4.592147e+00 4.592147e+00
## H.T.bank 4.574012e+00 4.574012e+00
## S.T.take 4.545539e+00 4.545539e+00
## S.T.day 4.501029e+00 4.501029e+00
## H.T.make 4.025030e+00 4.025030e+00
## H.T.time 3.968662e+00 3.968662e+00
## H.T.pictur 3.751942e+00 3.751942e+00
## H.T.billion 3.746828e+00 3.746828e+00
## H.T.X2014 2.938640e+00 2.938640e+00
## A.T.articl 2.744290e+00 2.744290e+00
## H.T.obama 2.631610e+00 2.631610e+00
## S.npnct16.log 2.586897e+00 2.586897e+00
## PubDate.last100.log 2.562314e+00 2.562314e+00
## S.npnct15.log 1.951937e+00 1.951937e+00
## A.T.first 1.686309e+00 1.686309e+00
## H.T.test 1.393050e+00 1.393050e+00
## A.npnct17.log 1.012142e+00 1.012142e+00
## S.npnct06.log 7.949129e-01 7.949129e-01
## S.T.new 6.136717e-01 6.136717e-01
## H.npnct30.log 5.296067e-01 5.296067e-01
## A.T.fashion 1.689321e-01 1.689321e-01
## H.T.springsumm 8.668136e-02 8.668136e-02
## H.T.deal 6.560495e-02 6.560495e-02
## S.npnct30.log 5.692475e-02 5.692475e-02
## H.npnct02.log 2.826878e-02 2.826878e-02
## S.npnct22.log 2.688801e-02 2.688801e-02
## S.npnct07.log 2.677639e-02 2.677639e-02
## A.T.presid 2.652248e-02 2.652248e-02
## S.T.presid 2.649780e-02 2.649780e-02
## S.npnct03.log 2.562697e-02 2.562697e-02
## S.has.year.colon 2.044323e-02 2.044323e-02
## H.npnct05.log 1.884000e-02 1.884000e-02
## S.npnct08.log 7.744037e-03 7.744037e-03
## S.npnct09.log 6.828413e-03 6.828413e-03
## A.npnct19.log 3.032825e-03 3.032825e-03
## A.npnct20.log 0.000000e+00 0.000000e+00
## A.has.http NA NA
## A.has.year.colon NA NA
## A.ndgts.log NA NA
## A.npnct01.log NA NA
## A.npnct02.log NA NA
## A.npnct03.log NA NA
## A.npnct04.log NA NA
## A.npnct05.log NA NA
## A.npnct06.log NA NA
## A.npnct07.log NA NA
## A.npnct08.log NA NA
## A.npnct09.log NA NA
## A.npnct10.log NA NA
## A.npnct11.log NA NA
## A.npnct12.log NA NA
## A.npnct13.log NA NA
## A.npnct15.log NA NA
## A.npnct16.log NA NA
## A.npnct18.log NA NA
## A.npnct22.log NA NA
## A.npnct23.log NA NA
## A.npnct24.log NA NA
## A.npnct25.log NA NA
## A.npnct26.log NA NA
## A.npnct27.log NA NA
## A.npnct28.log NA NA
## A.npnct29.log NA NA
## A.npnct30.log NA NA
## A.npnct31.log NA NA
## A.npnct32.log NA NA
## A.nuppr.log NA NA
## A.T.can NA NA
## A.T.day NA NA
## A.T.make NA NA
## A.T.new NA NA
## A.T.one NA NA
## A.T.said NA NA
## A.T.share NA NA
## A.T.show NA NA
## A.T.state NA NA
## A.T.take NA NA
## A.T.time NA NA
## clusterid NA NA
## H.has.http NA NA
## H.has.year.colon NA NA
## H.npnct03.log NA NA
## H.npnct04.log NA NA
## H.npnct06.log NA NA
## H.npnct08.log NA NA
## H.npnct10.log NA NA
## H.npnct11.log NA NA
## H.npnct15.log NA NA
## H.npnct18.log NA NA
## H.npnct19.log NA NA
## H.npnct20.log NA NA
## H.npnct22.log NA NA
## H.npnct23.log NA NA
## H.npnct24.log NA NA
## H.npnct25.log NA NA
## H.npnct26.log NA NA
## H.npnct27.log NA NA
## H.npnct28.log NA NA
## H.npnct29.log NA NA
## H.npnct31.log NA NA
## H.npnct32.log NA NA
## H.nwrds.unq.log NA NA
## H.T.daili NA NA
## H.T.morn NA NA
## H.T.today NA NA
## H.T.X2015 NA NA
## Popular NA NA
## Popular.fctr NA NA
## PubDate.last1 NA NA
## PubDate.last10 NA NA
## PubDate.last100 NA NA
## PubDate.month.fctr NA NA
## PubDate.POSIX NA NA
## PubDate.year.fctr NA NA
## PubDate.zoo NA NA
## S.has.http NA NA
## S.nchrs.log NA NA
## S.npnct02.log NA NA
## S.npnct05.log NA NA
## S.npnct10.log NA NA
## S.npnct11.log NA NA
## S.npnct14.log NA NA
## S.npnct17.log NA NA
## S.npnct18.log NA NA
## S.npnct19.log NA NA
## S.npnct20.log NA NA
## S.npnct21.log NA NA
## S.npnct23.log NA NA
## S.npnct24.log NA NA
## S.npnct25.log NA NA
## S.npnct26.log NA NA
## S.npnct27.log NA NA
## S.npnct28.log NA NA
## S.npnct29.log NA NA
## S.npnct31.log NA NA
## S.npnct32.log NA NA
## S.nwrds.log NA NA
## S.nwrds.unq.log NA NA
## S.T.articl NA NA
## S.T.compani NA NA
## S.T.fashion NA NA
## S.T.first NA NA
## S.T.intern NA NA
## S.T.newyork NA NA
## S.T.report NA NA
## S.T.week NA NA
## S.T.will NA NA
## S.T.year NA NA
## UniqueID NA NA
## WordCount NA NAglb_analytics_diag_plots(obs_df=glb_trnent_df, mdl_id=glb_fin_mdl_id,
prob_threshold=ifelse(glb_is_classification && glb_is_binomial,
glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"], NULL))## Warning in glb_analytics_diag_plots(obs_df = glb_trnent_df, mdl_id =
## glb_fin_mdl_id, : Limiting important feature scatter plots to 5 out of 100
## [1] "Min/Max Boundaries: "
## UniqueID Popular.fctr Popular.fctr.predict.Final.glm.prob
## 1507 1507 N 0.0004328477
## 6370 6370 Y 0.6234564223
## Popular.fctr.predict.Final.glm
## 1507 N
## 6370 Y
## Popular.fctr.predict.Final.glm.accurate
## 1507 TRUE
## 6370 TRUE
## Popular.fctr.predict.Final.glm.error .label
## 1507 0 1507
## 6370 0 6370
## [1] "Inaccurate: "
## UniqueID Popular.fctr Popular.fctr.predict.Final.glm.prob
## 1923 1923 Y 0.002113518
## 6101 6101 Y 0.002319721
## 2182 2182 Y 0.002863977
## 4020 4020 Y 0.003369097
## 4721 4721 Y 0.003844423
## 3113 3113 Y 0.004736358
## Popular.fctr.predict.Final.glm
## 1923 N
## 6101 N
## 2182 N
## 4020 N
## 4721 N
## 3113 N
## Popular.fctr.predict.Final.glm.accurate
## 1923 FALSE
## 6101 FALSE
## 2182 FALSE
## 4020 FALSE
## 4721 FALSE
## 3113 FALSE
## Popular.fctr.predict.Final.glm.error
## 1923 -0.2978865
## 6101 -0.2976803
## 2182 -0.2971360
## 4020 -0.2966309
## 4721 -0.2961556
## 3113 -0.2952636
## UniqueID Popular.fctr Popular.fctr.predict.Final.glm.prob
## 4020 4020 Y 0.003369097
## 3743 3743 Y 0.119953926
## 1096 1096 Y 0.153243330
## 1807 1807 Y 0.199739668
## 3196 3196 N 0.374767224
## 2708 2708 N 0.730538189
## Popular.fctr.predict.Final.glm
## 4020 N
## 3743 N
## 1096 N
## 1807 N
## 3196 Y
## 2708 Y
## Popular.fctr.predict.Final.glm.accurate
## 4020 FALSE
## 3743 FALSE
## 1096 FALSE
## 1807 FALSE
## 3196 FALSE
## 2708 FALSE
## Popular.fctr.predict.Final.glm.error
## 4020 -0.29663090
## 3743 -0.18004607
## 1096 -0.14675667
## 1807 -0.10026033
## 3196 0.07476722
## 2708 0.43053819
## UniqueID Popular.fctr Popular.fctr.predict.Final.glm.prob
## 1612 1612 N 0.9539704
## 1667 1667 N 0.9597500
## 4882 4882 N 0.9790763
## 770 770 N 0.9794789
## 1448 1448 N 0.9795772
## 59 59 N 0.9832044
## Popular.fctr.predict.Final.glm
## 1612 Y
## 1667 Y
## 4882 Y
## 770 Y
## 1448 Y
## 59 Y
## Popular.fctr.predict.Final.glm.accurate
## 1612 FALSE
## 1667 FALSE
## 4882 FALSE
## 770 FALSE
## 1448 FALSE
## 59 FALSE
## Popular.fctr.predict.Final.glm.error
## 1612 0.6539704
## 1667 0.6597500
## 4882 0.6790763
## 770 0.6794789
## 1448 0.6795772
## 59 0.6832044
dsp_feats_vctr <- c(NULL)
for(var in grep(".importance", names(glb_feats_df), fixed=TRUE, value=TRUE))
dsp_feats_vctr <- union(dsp_feats_vctr,
glb_feats_df[!is.na(glb_feats_df[, var]), "id"])
print(glb_trnent_df[glb_trnent_df$UniqueID %in% FN_OOB_ids,
grep(glb_rsp_var, names(glb_trnent_df), value=TRUE)])## Popular.fctr Popular.fctr.predict.Final.glm.prob
## 92 Y 0.046041464
## 693 Y 0.076887402
## 4020 Y 0.003369097
## 4721 Y 0.003844423
## Popular.fctr.predict.Final.glm
## 92 N
## 693 N
## 4020 N
## 4721 Nsav_entity_df <- glb_entity_df
print(setdiff(names(glb_trnent_df), names(glb_entity_df)))## [1] "Popular.fctr.predict.Final.glm.prob"
## [2] "Popular.fctr.predict.Final.glm"for (col in setdiff(names(glb_trnent_df), names(glb_entity_df)))
# Merge or cbind ?
glb_entity_df[glb_entity_df$.src == "Train", col] <- glb_trnent_df[, col]
print(setdiff(names(glb_fitent_df), names(glb_entity_df)))## character(0)print(setdiff(names(glb_OOBent_df), names(glb_entity_df)))## character(0)for (col in setdiff(names(glb_OOBent_df), names(glb_entity_df)))
# Merge or cbind ?
glb_entity_df[glb_entity_df$.lcn == "OOB", col] <- glb_OOBent_df[, col]
print(setdiff(names(glb_newent_df), names(glb_entity_df)))## character(0)if (glb_save_envir)
save(glb_feats_df, glb_entity_df,
#glb_trnent_df, glb_fitent_df, glb_OOBent_df, glb_newent_df,
glb_models_df, dsp_models_df, glb_models_lst, glb_model_type,
glb_sel_mdl, glb_sel_mdl_id,
glb_fin_mdl, glb_fin_mdl_id,
file=paste0(glb_out_pfx, "dsk.RData"))
replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.training.all.prediction","model.final")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
## 3.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: data.training.all.prediction
## 4.0000 5 0 1 1 1
## 4.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: model.final
## 5.0000 4 0 0 2 1
glb_chunks_df <- myadd_chunk(glb_chunks_df, "predict.data.new", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 15 fit.data.training 8 1 354.080 365.325 11.246
## 16 predict.data.new 9 0 365.326 NA NAStep 9.0: predict data new
# Compute final model predictions
glb_newent_df <- glb_get_predictions(glb_newent_df, mdl_id=glb_fin_mdl_id,
rsp_var_out=glb_rsp_var_out,
prob_threshold_def=ifelse(glb_is_classification && glb_is_binomial,
glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"], NULL))## Warning in glb_get_predictions(glb_newent_df, mdl_id = glb_fin_mdl_id,
## rsp_var_out = glb_rsp_var_out, : Using default probability threshold: 0.3glb_analytics_diag_plots(obs_df=glb_newent_df, mdl_id=glb_fin_mdl_id,
prob_threshold=ifelse(glb_is_classification && glb_is_binomial,
glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"], NULL))## Warning in glb_analytics_diag_plots(obs_df = glb_newent_df, mdl_id =
## glb_fin_mdl_id, : Limiting important feature scatter plots to 5 out of 100## Warning in min(x): no non-missing arguments to min; returning Inf## Warning in max(x): no non-missing arguments to max; returning -Inf## Warning: Removed 1870 rows containing missing values (geom_point).## Warning: Removed 1870 rows containing missing values (geom_point).
## Warning in min(x): no non-missing arguments to min; returning Inf## Warning in max(x): no non-missing arguments to max; returning -Inf## Warning: Removed 1870 rows containing missing values (geom_point).## Warning: Removed 1870 rows containing missing values (geom_point).
## Warning in min(x): no non-missing arguments to min; returning Inf## Warning in max(x): no non-missing arguments to max; returning -Inf## Warning: Removed 1870 rows containing missing values (geom_point).## Warning: Removed 1870 rows containing missing values (geom_point).
## Warning in min(x): no non-missing arguments to min; returning Inf## Warning in max(x): no non-missing arguments to max; returning -Inf## Warning: Removed 1870 rows containing missing values (geom_point).## Warning: Removed 1870 rows containing missing values (geom_point).
## Warning in min(x): no non-missing arguments to min; returning Inf## Warning in max(x): no non-missing arguments to max; returning -Inf## Warning: Removed 1870 rows containing missing values (geom_point).## Warning: Removed 1870 rows containing missing values (geom_point).
## [1] "Min/Max Boundaries: "
## UniqueID Popular.fctr Popular.fctr.predict.Final.glm.prob
## 6753 6753 <NA> 0.8089167723
## 7309 7309 <NA> 0.0002589035
## Popular.fctr.predict.Final.glm
## 6753 Y
## 7309 N
## Popular.fctr.predict.Final.glm.accurate
## 6753 NA
## 7309 NA
## Popular.fctr.predict.Final.glm.error .label
## 6753 0 6753
## 7309 0 7309
## [1] "Inaccurate: "
## UniqueID Popular.fctr Popular.fctr.predict.Final.glm.prob
## NA NA <NA> NA
## NA.1 NA <NA> NA
## NA.2 NA <NA> NA
## NA.3 NA <NA> NA
## NA.4 NA <NA> NA
## NA.5 NA <NA> NA
## Popular.fctr.predict.Final.glm
## NA <NA>
## NA.1 <NA>
## NA.2 <NA>
## NA.3 <NA>
## NA.4 <NA>
## NA.5 <NA>
## Popular.fctr.predict.Final.glm.accurate
## NA NA
## NA.1 NA
## NA.2 NA
## NA.3 NA
## NA.4 NA
## NA.5 NA
## Popular.fctr.predict.Final.glm.error
## NA NA
## NA.1 NA
## NA.2 NA
## NA.3 NA
## NA.4 NA
## NA.5 NA
## UniqueID Popular.fctr Popular.fctr.predict.Final.glm.prob
## NA.258 NA <NA> NA
## NA.373 NA <NA> NA
## NA.568 NA <NA> NA
## NA.1724 NA <NA> NA
## NA.1759 NA <NA> NA
## NA.1843 NA <NA> NA
## Popular.fctr.predict.Final.glm
## NA.258 <NA>
## NA.373 <NA>
## NA.568 <NA>
## NA.1724 <NA>
## NA.1759 <NA>
## NA.1843 <NA>
## Popular.fctr.predict.Final.glm.accurate
## NA.258 NA
## NA.373 NA
## NA.568 NA
## NA.1724 NA
## NA.1759 NA
## NA.1843 NA
## Popular.fctr.predict.Final.glm.error
## NA.258 NA
## NA.373 NA
## NA.568 NA
## NA.1724 NA
## NA.1759 NA
## NA.1843 NA
## UniqueID Popular.fctr Popular.fctr.predict.Final.glm.prob
## NA.1864 NA <NA> NA
## NA.1865 NA <NA> NA
## NA.1866 NA <NA> NA
## NA.1867 NA <NA> NA
## NA.1868 NA <NA> NA
## NA.1869 NA <NA> NA
## Popular.fctr.predict.Final.glm
## NA.1864 <NA>
## NA.1865 <NA>
## NA.1866 <NA>
## NA.1867 <NA>
## NA.1868 <NA>
## NA.1869 <NA>
## Popular.fctr.predict.Final.glm.accurate
## NA.1864 NA
## NA.1865 NA
## NA.1866 NA
## NA.1867 NA
## NA.1868 NA
## NA.1869 NA
## Popular.fctr.predict.Final.glm.error
## NA.1864 NA
## NA.1865 NA
## NA.1866 NA
## NA.1867 NA
## NA.1868 NA
## NA.1869 NA## Warning: Removed 1870 rows containing missing values (geom_point).
submit_df <- glb_newent_df[, c(glb_id_vars,
paste0(glb_rsp_var_out, glb_fin_mdl_id, ".prob"))]
names(submit_df)[2] <- "Probability1"
write.csv(submit_df,
paste0(gsub(".", "_", paste0(glb_out_pfx, glb_fin_mdl_id), fixed=TRUE),
"_submit.csv"), row.names=FALSE)
# print(orderBy(~ -max.auc.OOB, glb_models_df[, c("model_id",
# "max.auc.OOB", "max.Accuracy.OOB")]))
print(glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"])## [1] 0.3print(sprintf("glb_sel_mdl_id: %s", glb_sel_mdl_id))## [1] "glb_sel_mdl_id: Low.cor.X.glm"print(sprintf("glb_fin_mdl_id: %s", glb_fin_mdl_id))## [1] "glb_fin_mdl_id: Final.glm"print(dim(glb_fitent_df))## [1] 4475 224print(dsp_models_df)## model_id max.Accuracy.OOB max.auc.OOB max.Kappa.OOB
## 8 Low.cor.X.glm 0.9066602 0.9160473 0.6776204
## 9 All.X.glm 0.8954789 0.7769434 0.5961271
## 10 All.X.no.rnorm.rpart 0.8862421 0.7084504 0.5054039
## 1 MFO.myMFO_classfr 0.8327662 0.5000000 0.0000000
## 3 Max.cor.Y.cv.0.rpart 0.8327662 0.5000000 0.0000000
## 4 Max.cor.Y.cv.0.cp.0.rpart 0.8327662 0.5000000 0.0000000
## 5 Max.cor.Y.rpart 0.8327662 0.5000000 0.0000000
## 7 Interact.High.cor.Y.glm 0.7953330 0.7758607 0.3354449
## 6 Max.cor.Y.glm 0.7316480 0.7102060 0.2283681
## 2 Random.myrandom_classfr 0.1672338 0.4909227 0.0000000
## min.aic.fit opt.prob.threshold.OOB
## 8 2099.443 0.3
## 9 29689.534 0.9
## 10 NA 0.7
## 1 NA 0.5
## 3 NA 0.5
## 4 NA 0.5
## 5 NA 0.5
## 7 3399.630 0.3
## 6 3714.601 0.2
## 2 NA 0.1print(sprintf("%s OOB confusion matrix & accuracy: ", glb_sel_mdl_id))## [1] "Low.cor.X.glm OOB confusion matrix & accuracy: "print(t(confusionMatrix(glb_OOBent_df[, paste0(glb_rsp_var_out, glb_sel_mdl_id)],
glb_OOBent_df[, glb_rsp_var])$table))## Prediction
## Reference N Y
## N 1600 113
## Y 79 265tmp_OOBent_df <- glb_OOBent_df[, c("myCategory", predct_accurate_var_name)]
names(tmp_OOBent_df)[2] <- "accurate.OOB"
aOOB_ctgry_df <- mycreate_xtab_df(tmp_OOBent_df, names(tmp_OOBent_df))
aOOB_ctgry_df[is.na(aOOB_ctgry_df)] <- 0
aOOB_ctgry_df <- mutate(aOOB_ctgry_df,
.n.OOB = accurate.OOB.FALSE + accurate.OOB.TRUE,
max.accuracy.OOB = accurate.OOB.TRUE / .n.OOB)
intersect(names(glb_ctgry_df), names(aOOB_ctgry_df))## [1] "myCategory" ".n.OOB"glb_ctgry_df <- merge(glb_ctgry_df, aOOB_ctgry_df, all=TRUE)
print(orderBy(~-accurate.OOB.FALSE, glb_ctgry_df))## myCategory .n.OOB .n.Tst .freqRatio.Tst
## 1 ## 407 338 0.180748663
## 6 Business#Business Day#Dealbook 312 304 0.162566845
## 15 OpEd#Opinion# 154 164 0.087700535
## 18 Styles#U.S.# 54 62 0.033155080
## 9 Business#Technology# 114 113 0.060427807
## 16 Science#Health# 66 57 0.030481283
## 10 Culture#Arts# 225 244 0.130481283
## 8 Business#Crosswords/Games# 40 42 0.022459893
## 13 Metro#N.Y. / Region# 60 67 0.035828877
## 3 #Opinion#Room For Debate 21 24 0.012834225
## 4 #Opinion#The Public Editor 10 10 0.005347594
## 7 Business#Business Day#Small Business 45 42 0.022459893
## 20 TStyle## 221 105 0.056149733
## 17 Styles##Fashion 41 15 0.008021390
## 2 #Multimedia# 42 52 0.027807487
## 5 #U.S.#Education 93 90 0.048128342
## 11 Foreign#World# 47 47 0.025133690
## 12 Foreign#World#Asia Pacific 61 56 0.029946524
## 14 myOther 13 3 0.001604278
## 19 Travel#Travel# 31 35 0.018716578
## .freqRatio.OOB accurate.OOB.FALSE accurate.OOB.TRUE max.accuracy.OOB
## 1 0.197860963 37 370 0.9090909
## 6 0.151677200 30 282 0.9038462
## 15 0.074866310 30 124 0.8051948
## 18 0.026251823 21 33 0.6111111
## 9 0.055420515 20 94 0.8245614
## 16 0.032085561 18 48 0.7272727
## 10 0.109382596 12 213 0.9466667
## 8 0.019445795 8 32 0.8000000
## 13 0.029168692 5 55 0.9166667
## 3 0.010209042 3 18 0.8571429
## 4 0.004861449 3 7 0.7000000
## 7 0.021876519 2 43 0.9555556
## 20 0.107438017 2 219 0.9909502
## 17 0.019931940 1 40 0.9756098
## 2 0.020418085 0 42 1.0000000
## 5 0.045211473 0 93 1.0000000
## 11 0.022848809 0 47 1.0000000
## 12 0.029654837 0 61 1.0000000
## 14 0.006319883 0 13 1.0000000
## 19 0.015070491 0 31 1.0000000dsp_myCategory_conf_mtrx <- function(myCategory) {
print(sprintf("%s OOB::myCategory=%s confusion matrix & accuracy: ",
glb_sel_mdl_id, myCategory))
print(t(confusionMatrix(
glb_OOBent_df[glb_OOBent_df$myCategory == myCategory,
paste0(glb_rsp_var_out, glb_sel_mdl_id)],
glb_OOBent_df[glb_OOBent_df$myCategory == myCategory, glb_rsp_var])$table))
print(sum(glb_OOBent_df[glb_OOBent_df$myCategory == myCategory,
predct_accurate_var_name]) /
nrow(glb_OOBent_df[glb_OOBent_df$myCategory == myCategory, ]))
err_ids <- glb_OOBent_df[(glb_OOBent_df$myCategory == myCategory) &
(!glb_OOBent_df[, predct_accurate_var_name]), glb_id_vars]
OOBerr_df <- glb_OOBent_df[(glb_OOBent_df$UniqueID %in% err_ids) &
(glb_OOBent_df$Popular == 1),
c("clusterid", "Headline", "Popular")]
print(sprintf("%s OOB::myCategory=%s FN errors: %d", glb_sel_mdl_id, myCategory,
nrow(OOBerr_df)))
print(OOBerr_df)
}
dsp_myCategory_conf_mtrx(myCategory="Science#Health#")## [1] "Low.cor.X.glm OOB::myCategory=Science#Health# confusion matrix & accuracy: "
## Prediction
## Reference N Y
## N 6 17
## Y 1 42
## [1] 0.7272727
## [1] "Low.cor.X.glm OOB::myCategory=Science#Health# FN errors: 1"
## clusterid
## 5701 1
## Headline Popular
## 5701 Vegetarian Thanksgiving: Caramelized Onion and Fennel Risotto 1print("FN_OOB_ids:")## [1] "FN_OOB_ids:"print(glb_OOBent_df[glb_OOBent_df$UniqueID %in% FN_OOB_ids,
grep(glb_rsp_var, names(glb_OOBent_df), value=TRUE)])## [1] Popular.fctr
## [2] Popular.fctr.predict.Low.cor.X.glm.prob
## [3] Popular.fctr.predict.Low.cor.X.glm
## [4] Popular.fctr.predict.Low.cor.X.glm.accurate
## <0 rows> (or 0-length row.names)print(glb_OOBent_df[glb_OOBent_df$UniqueID %in% FN_OOB_ids,
glb_txt_vars])## [1] Headline Snippet Abstract
## <0 rows> (or 0-length row.names)print(dsp_vctr <- colSums(glb_OOBent_df[glb_OOBent_df$UniqueID %in% FN_OOB_ids,
setdiff(grep("[HSA].", names(glb_OOBent_df), value=TRUE),
union(myfind_chr_cols_df(glb_OOBent_df),
grep(".fctr", names(glb_OOBent_df), fixed=TRUE, value=TRUE)))]))## PubDate.POSIX H.T.X2014 H.T.X2015 H.T.art
## 0 0 0 0
## H.T.bank H.T.big H.T.billion H.T.busi
## 0 0 0 0
## H.T.china H.T.daili H.T.day H.T.deal
## 0 0 0 0
## H.T.fashion H.T.first H.T.make H.T.morn
## 0 0 0 0
## H.T.new H.T.news H.T.newyork H.T.obama
## 0 0 0 0
## H.T.pictur H.T.polit H.T.report H.T.say
## 0 0 0 0
## H.T.springsumm H.T.take H.T.test H.T.time
## 0 0 0 0
## H.T.today H.T.week H.has.http H.has.ebola
## 0 0 0 0
## H.nwrds.log H.nwrds.unq.log H.nchrs.log H.nuppr.log
## 0 0 0 0
## H.ndgts.log H.npnct01.log H.npnct02.log H.npnct03.log
## 0 0 0 0
## H.npnct04.log H.npnct05.log H.npnct06.log H.npnct07.log
## 0 0 0 0
## H.npnct08.log H.npnct09.log H.npnct10.log H.npnct11.log
## 0 0 0 0
## H.npnct12.log H.npnct13.log H.npnct14.log H.npnct15.log
## 0 0 0 0
## H.npnct16.log H.npnct17.log H.npnct18.log H.npnct19.log
## 0 0 0 0
## H.npnct20.log H.npnct21.log H.npnct22.log H.npnct23.log
## 0 0 0 0
## H.npnct24.log H.npnct25.log H.npnct26.log H.npnct27.log
## 0 0 0 0
## H.npnct28.log H.npnct29.log H.npnct30.log H.npnct31.log
## 0 0 0 0
## H.npnct32.log H.has.year.colon S.T.articl S.T.can
## 0 0 0 0
## S.T.compani S.T.day S.T.fashion S.T.first
## 0 0 0 0
## S.T.intern S.T.make S.T.new S.T.newyork
## 0 0 0 0
## S.T.one S.T.presid S.T.report S.T.said
## 0 0 0 0
## S.T.share S.T.show S.T.state S.T.take
## 0 0 0 0
## S.T.time S.T.week S.T.will S.T.year
## 0 0 0 0
## S.has.http S.nwrds.log S.nwrds.unq.log S.nchrs.log
## 0 0 0 0
## S.nuppr.log S.ndgts.log S.npnct01.log S.npnct02.log
## 0 0 0 0
## S.npnct03.log S.npnct04.log S.npnct05.log S.npnct06.log
## 0 0 0 0
## S.npnct07.log S.npnct08.log S.npnct09.log S.npnct10.log
## 0 0 0 0
## S.npnct11.log S.npnct12.log S.npnct13.log S.npnct14.log
## 0 0 0 0
## S.npnct15.log S.npnct16.log S.npnct17.log S.npnct18.log
## 0 0 0 0
## S.npnct19.log S.npnct20.log S.npnct21.log S.npnct22.log
## 0 0 0 0
## S.npnct23.log S.npnct24.log S.npnct25.log S.npnct26.log
## 0 0 0 0
## S.npnct27.log S.npnct28.log S.npnct29.log S.npnct30.log
## 0 0 0 0
## S.npnct31.log S.npnct32.log S.has.year.colon A.T.articl
## 0 0 0 0
## A.T.can A.T.compani A.T.day A.T.fashion
## 0 0 0 0
## A.T.first A.T.intern A.T.make A.T.new
## 0 0 0 0
## A.T.newyork A.T.one A.T.presid A.T.report
## 0 0 0 0
## A.T.said A.T.share A.T.show A.T.state
## 0 0 0 0
## A.T.take A.T.time A.T.week A.T.will
## 0 0 0 0
## A.T.year A.has.http A.nwrds.log A.nwrds.unq.log
## 0 0 0 0
## A.nchrs.log A.nuppr.log A.ndgts.log A.npnct01.log
## 0 0 0 0
## A.npnct02.log A.npnct03.log A.npnct04.log A.npnct05.log
## 0 0 0 0
## A.npnct06.log A.npnct07.log A.npnct08.log A.npnct09.log
## 0 0 0 0
## A.npnct10.log A.npnct11.log A.npnct12.log A.npnct13.log
## 0 0 0 0
## A.npnct14.log A.npnct15.log A.npnct16.log A.npnct17.log
## 0 0 0 0
## A.npnct18.log A.npnct19.log A.npnct20.log A.npnct21.log
## 0 0 0 0
## A.npnct22.log A.npnct23.log A.npnct24.log A.npnct25.log
## 0 0 0 0
## A.npnct26.log A.npnct27.log A.npnct28.log A.npnct29.log
## 0 0 0 0
## A.npnct30.log A.npnct31.log A.npnct32.log A.has.year.colon
## 0 0 0 0dsp_hdlpfx_results <- function(hdlpfx) {
print(hdlpfx)
print(glb_OOBent_df[glb_OOBent_df$Headline.pfx %in% c(hdlpfx),
grep(glb_rsp_var, names(glb_OOBent_df), value=TRUE)])
print(glb_newent_df[glb_newent_df$Headline.pfx %in% c(hdlpfx),
grep(glb_rsp_var, names(glb_newent_df), value=TRUE)])
print(dsp_vctr <- colSums(glb_newent_df[glb_newent_df$Headline.pfx %in% c(hdlpfx),
setdiff(grep("[HSA]\\.", names(glb_newent_df), value=TRUE),
union(myfind_chr_cols_df(glb_newent_df),
grep(".fctr", names(glb_newent_df), fixed=TRUE, value=TRUE)))]))
print(dsp_vctr <- dsp_vctr[dsp_vctr != 0])
print(glb_newent_df[glb_newent_df$Headline.pfx %in% c(hdlpfx),
union(names(dsp_vctr), myfind_chr_cols_df(glb_newent_df))])
}
#dsp_hdlpfx_results(hdlpfx="Ask Well::")
# print("myMisc::|OpEd|blank|blank|1:")
# print(glb_OOBent_df[glb_OOBent_df$UniqueID %in% c(6446),
# grep(glb_rsp_var, names(glb_OOBent_df), value=TRUE)])
# print(glb_OOBent_df[glb_OOBent_df$UniqueID %in% FN_OOB_ids,
# c("WordCount", "WordCount.log", "myMultimedia",
# "NewsDesk", "SectionName", "SubsectionName")])
# print(mycreate_sqlxtab_df(glb_entity_df[sel_obs(Headline.contains="[Vv]ideo"), ],
# c(glb_rsp_var, "myMultimedia")))
# dsp_chisq.test(Headline.contains="[Vi]deo")
# print(glb_entity_df[sel_obs(Headline.contains="[Vv]ideo"),
# c(glb_rsp_var, "Popular", "myMultimedia", "Headline")])
# print(glb_entity_df[sel_obs(Headline.contains="[Ee]bola", Popular=1),
# c(glb_rsp_var, "Popular", "myMultimedia", "Headline",
# "NewsDesk", "SectionName", "SubsectionName")])
# print(subset(glb_feats_df, !is.na(importance))[,
# c("is.ConditionalX.y",
# grep("importance", names(glb_feats_df), fixed=TRUE, value=TRUE))])
# print(subset(glb_feats_df, is.ConditionalX.y & is.na(importance))[,
# c("is.ConditionalX.y",
# grep("importance", names(glb_feats_df), fixed=TRUE, value=TRUE))])
print(subset(glb_feats_df, !is.na(importance))[,
c("zeroVar", "nzv", "myNearZV",
grep("importance", names(glb_feats_df), fixed=TRUE, value=TRUE))])## zeroVar nzv myNearZV importance
## WordCount.log FALSE FALSE FALSE 1.000000e+02
## myCategory.fctr FALSE FALSE FALSE 7.312143e+01
## H.npnct21.log FALSE FALSE FALSE 3.777302e+01
## A.npnct21.log FALSE FALSE FALSE 3.475205e+01
## S.nuppr.log FALSE FALSE FALSE 3.238555e+01
## H.nchrs.log FALSE FALSE FALSE 3.201395e+01
## A.npnct14.log FALSE FALSE FALSE 2.764418e+01
## H.npnct09.log FALSE TRUE FALSE 2.188969e+01
## PubDate.wkday.fctr FALSE FALSE FALSE 2.165613e+01
## H.nuppr.log FALSE FALSE FALSE 2.131190e+01
## H.T.polit FALSE TRUE FALSE 2.052535e+01
## A.T.newyork FALSE TRUE FALSE 1.961261e+01
## A.T.report FALSE TRUE FALSE 1.938341e+01
## H.npnct12.log FALSE FALSE FALSE 1.796841e+01
## S.ndgts.log FALSE FALSE FALSE 1.621363e+01
## A.nwrds.unq.log FALSE FALSE FALSE 1.619197e+01
## S.T.said FALSE TRUE FALSE 1.520654e+01
## S.T.make FALSE TRUE FALSE 1.512129e+01
## H.npnct17.log FALSE TRUE FALSE 1.480733e+01
## PubDate.last10.log FALSE FALSE FALSE 1.453274e+01
## S.T.state FALSE TRUE FALSE 1.414282e+01
## S.T.one FALSE TRUE FALSE 1.375899e+01
## S.npnct04.log FALSE TRUE FALSE 1.373898e+01
## S.T.share FALSE TRUE FALSE 1.314940e+01
## PubDate.hour.fctr FALSE FALSE FALSE 1.314384e+01
## H.T.report FALSE TRUE FALSE 1.299102e+01
## PubDate.second.fctr FALSE FALSE FALSE 1.273156e+01
## H.T.newyork FALSE TRUE FALSE 1.269274e+01
## S.T.can FALSE TRUE FALSE 1.228026e+01
## H.T.say FALSE TRUE FALSE 1.222772e+01
## A.T.week FALSE TRUE FALSE 1.168514e+01
## H.T.fashion FALSE TRUE FALSE 1.151035e+01
## H.npnct16.log FALSE FALSE FALSE 1.136051e+01
## A.T.will FALSE TRUE FALSE 1.129252e+01
## S.T.show FALSE TRUE FALSE 1.089884e+01
## PubDate.minute.fctr FALSE FALSE FALSE 1.041575e+01
## H.ndgts.log FALSE FALSE FALSE 1.035074e+01
## S.npnct12.log FALSE FALSE FALSE 9.930532e+00
## H.npnct07.log FALSE FALSE FALSE 9.854654e+00
## A.nwrds.log FALSE FALSE FALSE 9.770236e+00
## H.T.art FALSE TRUE FALSE 9.623245e+00
## H.npnct01.log FALSE TRUE FALSE 9.509217e+00
## .rnorm FALSE FALSE FALSE 9.238260e+00
## S.T.time FALSE TRUE FALSE 9.237301e+00
## H.npnct13.log FALSE FALSE FALSE 9.109144e+00
## S.npnct01.log FALSE TRUE FALSE 8.749827e+00
## A.T.intern FALSE TRUE FALSE 8.472657e+00
## A.T.compani FALSE TRUE FALSE 8.369682e+00
## H.T.new FALSE TRUE FALSE 8.270503e+00
## A.nchrs.log FALSE FALSE FALSE 8.248780e+00
## PubDate.last1.log FALSE FALSE FALSE 8.146784e+00
## H.npnct14.log FALSE TRUE FALSE 8.031911e+00
## H.T.first FALSE TRUE FALSE 7.798393e+00
## PubDate.date.fctr FALSE FALSE FALSE 7.300560e+00
## H.T.busi FALSE TRUE FALSE 7.168757e+00
## S.npnct13.log FALSE FALSE FALSE 6.710517e+00
## H.T.day FALSE TRUE FALSE 6.378120e+00
## H.T.china FALSE TRUE FALSE 6.350450e+00
## A.T.year FALSE TRUE FALSE 6.178982e+00
## H.nwrds.log FALSE FALSE FALSE 6.165115e+00
## PubDate.wkend FALSE FALSE FALSE 6.020520e+00
## H.T.big FALSE TRUE FALSE 6.007063e+00
## H.T.news FALSE TRUE FALSE 5.524332e+00
## H.T.week FALSE TRUE FALSE 5.347019e+00
## H.has.ebola FALSE TRUE FALSE 4.853552e+00
## H.T.take FALSE TRUE FALSE 4.592147e+00
## H.T.bank FALSE TRUE FALSE 4.574012e+00
## S.T.take FALSE TRUE FALSE 4.545539e+00
## S.T.day FALSE TRUE FALSE 4.501029e+00
## H.T.make FALSE TRUE FALSE 4.025030e+00
## H.T.time FALSE TRUE FALSE 3.968662e+00
## H.T.pictur FALSE TRUE FALSE 3.751942e+00
## H.T.billion FALSE TRUE FALSE 3.746828e+00
## H.T.X2014 FALSE TRUE FALSE 2.938640e+00
## A.T.articl FALSE TRUE FALSE 2.744290e+00
## H.T.obama FALSE TRUE FALSE 2.631610e+00
## S.npnct16.log FALSE FALSE FALSE 2.586897e+00
## PubDate.last100.log FALSE FALSE FALSE 2.562314e+00
## S.npnct15.log FALSE TRUE FALSE 1.951937e+00
## A.T.first FALSE TRUE FALSE 1.686309e+00
## H.T.test FALSE TRUE FALSE 1.393050e+00
## A.npnct17.log FALSE TRUE FALSE 1.012142e+00
## S.npnct06.log FALSE TRUE FALSE 7.949129e-01
## S.T.new FALSE TRUE FALSE 6.136717e-01
## H.npnct30.log FALSE TRUE FALSE 5.296067e-01
## A.T.fashion FALSE TRUE FALSE 1.689321e-01
## H.T.springsumm FALSE TRUE FALSE 8.668136e-02
## H.T.deal FALSE TRUE FALSE 6.560495e-02
## S.npnct30.log FALSE TRUE FALSE 5.692475e-02
## H.npnct02.log FALSE TRUE FALSE 2.826878e-02
## S.npnct22.log FALSE TRUE FALSE 2.688801e-02
## S.npnct07.log FALSE TRUE FALSE 2.677639e-02
## A.T.presid FALSE TRUE FALSE 2.652248e-02
## S.T.presid FALSE TRUE FALSE 2.649780e-02
## S.npnct03.log FALSE TRUE FALSE 2.562697e-02
## S.has.year.colon FALSE TRUE FALSE 2.044323e-02
## H.npnct05.log FALSE TRUE FALSE 1.884000e-02
## S.npnct08.log FALSE TRUE FALSE 7.744037e-03
## S.npnct09.log FALSE TRUE FALSE 6.828413e-03
## A.npnct19.log FALSE TRUE FALSE 3.032825e-03
## A.npnct20.log FALSE TRUE FALSE 0.000000e+00
## Low.cor.X.glm.importance Final.glm.importance
## WordCount.log 1.000000e+02 1.000000e+02
## myCategory.fctr 7.312143e+01 7.312143e+01
## H.npnct21.log 3.777302e+01 3.777302e+01
## A.npnct21.log 3.475205e+01 3.475205e+01
## S.nuppr.log 3.238555e+01 3.238555e+01
## H.nchrs.log 3.201395e+01 3.201395e+01
## A.npnct14.log 2.764418e+01 2.764418e+01
## H.npnct09.log 2.188969e+01 2.188969e+01
## PubDate.wkday.fctr 2.165613e+01 2.165613e+01
## H.nuppr.log 2.131190e+01 2.131190e+01
## H.T.polit 2.052535e+01 2.052535e+01
## A.T.newyork 1.961261e+01 1.961261e+01
## A.T.report 1.938341e+01 1.938341e+01
## H.npnct12.log 1.796841e+01 1.796841e+01
## S.ndgts.log 1.621363e+01 1.621363e+01
## A.nwrds.unq.log 1.619197e+01 1.619197e+01
## S.T.said 1.520654e+01 1.520654e+01
## S.T.make 1.512129e+01 1.512129e+01
## H.npnct17.log 1.480733e+01 1.480733e+01
## PubDate.last10.log 1.453274e+01 1.453274e+01
## S.T.state 1.414282e+01 1.414282e+01
## S.T.one 1.375899e+01 1.375899e+01
## S.npnct04.log 1.373898e+01 1.373898e+01
## S.T.share 1.314940e+01 1.314940e+01
## PubDate.hour.fctr 1.314384e+01 1.314384e+01
## H.T.report 1.299102e+01 1.299102e+01
## PubDate.second.fctr 1.273156e+01 1.273156e+01
## H.T.newyork 1.269274e+01 1.269274e+01
## S.T.can 1.228026e+01 1.228026e+01
## H.T.say 1.222772e+01 1.222772e+01
## A.T.week 1.168514e+01 1.168514e+01
## H.T.fashion 1.151035e+01 1.151035e+01
## H.npnct16.log 1.136051e+01 1.136051e+01
## A.T.will 1.129252e+01 1.129252e+01
## S.T.show 1.089884e+01 1.089884e+01
## PubDate.minute.fctr 1.041575e+01 1.041575e+01
## H.ndgts.log 1.035074e+01 1.035074e+01
## S.npnct12.log 9.930532e+00 9.930532e+00
## H.npnct07.log 9.854654e+00 9.854654e+00
## A.nwrds.log 9.770236e+00 9.770236e+00
## H.T.art 9.623245e+00 9.623245e+00
## H.npnct01.log 9.509217e+00 9.509217e+00
## .rnorm 9.238260e+00 9.238260e+00
## S.T.time 9.237301e+00 9.237301e+00
## H.npnct13.log 9.109144e+00 9.109144e+00
## S.npnct01.log 8.749827e+00 8.749827e+00
## A.T.intern 8.472657e+00 8.472657e+00
## A.T.compani 8.369682e+00 8.369682e+00
## H.T.new 8.270503e+00 8.270503e+00
## A.nchrs.log 8.248780e+00 8.248780e+00
## PubDate.last1.log 8.146784e+00 8.146784e+00
## H.npnct14.log 8.031911e+00 8.031911e+00
## H.T.first 7.798393e+00 7.798393e+00
## PubDate.date.fctr 7.300560e+00 7.300560e+00
## H.T.busi 7.168757e+00 7.168757e+00
## S.npnct13.log 6.710517e+00 6.710517e+00
## H.T.day 6.378120e+00 6.378120e+00
## H.T.china 6.350450e+00 6.350450e+00
## A.T.year 6.178982e+00 6.178982e+00
## H.nwrds.log 6.165115e+00 6.165115e+00
## PubDate.wkend 6.020520e+00 6.020520e+00
## H.T.big 6.007063e+00 6.007063e+00
## H.T.news 5.524332e+00 5.524332e+00
## H.T.week 5.347019e+00 5.347019e+00
## H.has.ebola 4.853552e+00 4.853552e+00
## H.T.take 4.592147e+00 4.592147e+00
## H.T.bank 4.574012e+00 4.574012e+00
## S.T.take 4.545539e+00 4.545539e+00
## S.T.day 4.501029e+00 4.501029e+00
## H.T.make 4.025030e+00 4.025030e+00
## H.T.time 3.968662e+00 3.968662e+00
## H.T.pictur 3.751942e+00 3.751942e+00
## H.T.billion 3.746828e+00 3.746828e+00
## H.T.X2014 2.938640e+00 2.938640e+00
## A.T.articl 2.744290e+00 2.744290e+00
## H.T.obama 2.631610e+00 2.631610e+00
## S.npnct16.log 2.586897e+00 2.586897e+00
## PubDate.last100.log 2.562314e+00 2.562314e+00
## S.npnct15.log 1.951937e+00 1.951937e+00
## A.T.first 1.686309e+00 1.686309e+00
## H.T.test 1.393050e+00 1.393050e+00
## A.npnct17.log 1.012142e+00 1.012142e+00
## S.npnct06.log 7.949129e-01 7.949129e-01
## S.T.new 6.136717e-01 6.136717e-01
## H.npnct30.log 5.296067e-01 5.296067e-01
## A.T.fashion 1.689321e-01 1.689321e-01
## H.T.springsumm 8.668136e-02 8.668136e-02
## H.T.deal 6.560495e-02 6.560495e-02
## S.npnct30.log 5.692475e-02 5.692475e-02
## H.npnct02.log 2.826878e-02 2.826878e-02
## S.npnct22.log 2.688801e-02 2.688801e-02
## S.npnct07.log 2.677639e-02 2.677639e-02
## A.T.presid 2.652248e-02 2.652248e-02
## S.T.presid 2.649780e-02 2.649780e-02
## S.npnct03.log 2.562697e-02 2.562697e-02
## S.has.year.colon 2.044323e-02 2.044323e-02
## H.npnct05.log 1.884000e-02 1.884000e-02
## S.npnct08.log 7.744037e-03 7.744037e-03
## S.npnct09.log 6.828413e-03 6.828413e-03
## A.npnct19.log 3.032825e-03 3.032825e-03
## A.npnct20.log 0.000000e+00 0.000000e+00print(subset(glb_feats_df, is.na(importance))[,
c("zeroVar", "nzv", "myNearZV",
grep("importance", names(glb_feats_df), fixed=TRUE, value=TRUE))])## zeroVar nzv myNearZV importance
## A.has.http FALSE TRUE FALSE NA
## A.has.year.colon FALSE TRUE FALSE NA
## A.ndgts.log FALSE FALSE FALSE NA
## A.npnct01.log FALSE TRUE FALSE NA
## A.npnct02.log FALSE TRUE FALSE NA
## A.npnct03.log FALSE TRUE FALSE NA
## A.npnct04.log FALSE TRUE FALSE NA
## A.npnct05.log TRUE TRUE TRUE NA
## A.npnct06.log FALSE TRUE FALSE NA
## A.npnct07.log FALSE TRUE FALSE NA
## A.npnct08.log FALSE TRUE FALSE NA
## A.npnct09.log FALSE TRUE FALSE NA
## A.npnct10.log TRUE TRUE TRUE NA
## A.npnct11.log FALSE TRUE TRUE NA
## A.npnct12.log FALSE FALSE FALSE NA
## A.npnct13.log FALSE FALSE FALSE NA
## A.npnct15.log FALSE TRUE FALSE NA
## A.npnct16.log FALSE FALSE FALSE NA
## A.npnct18.log FALSE TRUE FALSE NA
## A.npnct22.log FALSE TRUE FALSE NA
## A.npnct23.log FALSE TRUE TRUE NA
## A.npnct24.log TRUE TRUE TRUE NA
## A.npnct25.log FALSE TRUE TRUE NA
## A.npnct26.log TRUE TRUE TRUE NA
## A.npnct27.log FALSE TRUE TRUE NA
## A.npnct28.log TRUE TRUE TRUE NA
## A.npnct29.log TRUE TRUE TRUE NA
## A.npnct30.log FALSE TRUE FALSE NA
## A.npnct31.log TRUE TRUE TRUE NA
## A.npnct32.log TRUE TRUE TRUE NA
## A.nuppr.log FALSE FALSE FALSE NA
## A.T.can FALSE TRUE FALSE NA
## A.T.day FALSE TRUE FALSE NA
## A.T.make FALSE TRUE FALSE NA
## A.T.new FALSE TRUE FALSE NA
## A.T.one FALSE TRUE FALSE NA
## A.T.said FALSE TRUE FALSE NA
## A.T.share FALSE TRUE FALSE NA
## A.T.show FALSE TRUE FALSE NA
## A.T.state FALSE TRUE FALSE NA
## A.T.take FALSE TRUE FALSE NA
## A.T.time FALSE TRUE FALSE NA
## clusterid TRUE TRUE TRUE NA
## H.has.http TRUE TRUE TRUE NA
## H.has.year.colon FALSE TRUE FALSE NA
## H.npnct03.log FALSE TRUE TRUE NA
## H.npnct04.log FALSE TRUE FALSE NA
## H.npnct06.log FALSE TRUE FALSE NA
## H.npnct08.log FALSE TRUE FALSE NA
## H.npnct10.log TRUE TRUE TRUE NA
## H.npnct11.log FALSE TRUE TRUE NA
## H.npnct15.log FALSE TRUE FALSE NA
## H.npnct18.log TRUE TRUE TRUE NA
## H.npnct19.log TRUE TRUE TRUE NA
## H.npnct20.log TRUE TRUE TRUE NA
## H.npnct22.log FALSE TRUE TRUE NA
## H.npnct23.log TRUE TRUE TRUE NA
## H.npnct24.log TRUE TRUE TRUE NA
## H.npnct25.log TRUE TRUE TRUE NA
## H.npnct26.log TRUE TRUE TRUE NA
## H.npnct27.log TRUE TRUE TRUE NA
## H.npnct28.log TRUE TRUE TRUE NA
## H.npnct29.log TRUE TRUE TRUE NA
## H.npnct31.log TRUE TRUE TRUE NA
## H.npnct32.log TRUE TRUE TRUE NA
## H.nwrds.unq.log FALSE FALSE FALSE NA
## H.T.daili FALSE TRUE FALSE NA
## H.T.morn FALSE TRUE FALSE NA
## H.T.today FALSE TRUE FALSE NA
## H.T.X2015 FALSE TRUE FALSE NA
## Popular FALSE FALSE FALSE NA
## Popular.fctr NA NA NA NA
## PubDate.last1 FALSE FALSE FALSE NA
## PubDate.last10 FALSE FALSE FALSE NA
## PubDate.last100 FALSE FALSE FALSE NA
## PubDate.month.fctr FALSE FALSE FALSE NA
## PubDate.POSIX FALSE FALSE FALSE NA
## PubDate.year.fctr TRUE TRUE TRUE NA
## PubDate.zoo FALSE FALSE FALSE NA
## S.has.http TRUE TRUE TRUE NA
## S.nchrs.log FALSE FALSE FALSE NA
## S.npnct02.log FALSE TRUE TRUE NA
## S.npnct05.log TRUE TRUE TRUE NA
## S.npnct10.log TRUE TRUE TRUE NA
## S.npnct11.log FALSE TRUE TRUE NA
## S.npnct14.log FALSE FALSE FALSE NA
## S.npnct17.log FALSE TRUE FALSE NA
## S.npnct18.log TRUE TRUE TRUE NA
## S.npnct19.log TRUE TRUE TRUE NA
## S.npnct20.log TRUE TRUE TRUE NA
## S.npnct21.log FALSE FALSE FALSE NA
## S.npnct23.log FALSE TRUE TRUE NA
## S.npnct24.log TRUE TRUE TRUE NA
## S.npnct25.log FALSE TRUE TRUE NA
## S.npnct26.log TRUE TRUE TRUE NA
## S.npnct27.log TRUE TRUE TRUE NA
## S.npnct28.log TRUE TRUE TRUE NA
## S.npnct29.log TRUE TRUE TRUE NA
## S.npnct31.log TRUE TRUE TRUE NA
## S.npnct32.log TRUE TRUE TRUE NA
## S.nwrds.log FALSE FALSE FALSE NA
## S.nwrds.unq.log FALSE FALSE FALSE NA
## S.T.articl FALSE TRUE FALSE NA
## S.T.compani FALSE TRUE FALSE NA
## S.T.fashion FALSE TRUE FALSE NA
## S.T.first FALSE TRUE FALSE NA
## S.T.intern FALSE TRUE FALSE NA
## S.T.newyork FALSE TRUE FALSE NA
## S.T.report FALSE TRUE FALSE NA
## S.T.week FALSE TRUE FALSE NA
## S.T.will FALSE TRUE FALSE NA
## S.T.year FALSE TRUE FALSE NA
## UniqueID FALSE FALSE FALSE NA
## WordCount FALSE FALSE FALSE NA
## Low.cor.X.glm.importance Final.glm.importance
## A.has.http NA NA
## A.has.year.colon NA NA
## A.ndgts.log NA NA
## A.npnct01.log NA NA
## A.npnct02.log NA NA
## A.npnct03.log NA NA
## A.npnct04.log NA NA
## A.npnct05.log NA NA
## A.npnct06.log NA NA
## A.npnct07.log NA NA
## A.npnct08.log NA NA
## A.npnct09.log NA NA
## A.npnct10.log NA NA
## A.npnct11.log NA NA
## A.npnct12.log NA NA
## A.npnct13.log NA NA
## A.npnct15.log NA NA
## A.npnct16.log NA NA
## A.npnct18.log NA NA
## A.npnct22.log NA NA
## A.npnct23.log NA NA
## A.npnct24.log NA NA
## A.npnct25.log NA NA
## A.npnct26.log NA NA
## A.npnct27.log NA NA
## A.npnct28.log NA NA
## A.npnct29.log NA NA
## A.npnct30.log NA NA
## A.npnct31.log NA NA
## A.npnct32.log NA NA
## A.nuppr.log NA NA
## A.T.can NA NA
## A.T.day NA NA
## A.T.make NA NA
## A.T.new NA NA
## A.T.one NA NA
## A.T.said NA NA
## A.T.share NA NA
## A.T.show NA NA
## A.T.state NA NA
## A.T.take NA NA
## A.T.time NA NA
## clusterid NA NA
## H.has.http NA NA
## H.has.year.colon NA NA
## H.npnct03.log NA NA
## H.npnct04.log NA NA
## H.npnct06.log NA NA
## H.npnct08.log NA NA
## H.npnct10.log NA NA
## H.npnct11.log NA NA
## H.npnct15.log NA NA
## H.npnct18.log NA NA
## H.npnct19.log NA NA
## H.npnct20.log NA NA
## H.npnct22.log NA NA
## H.npnct23.log NA NA
## H.npnct24.log NA NA
## H.npnct25.log NA NA
## H.npnct26.log NA NA
## H.npnct27.log NA NA
## H.npnct28.log NA NA
## H.npnct29.log NA NA
## H.npnct31.log NA NA
## H.npnct32.log NA NA
## H.nwrds.unq.log NA NA
## H.T.daili NA NA
## H.T.morn NA NA
## H.T.today NA NA
## H.T.X2015 NA NA
## Popular NA NA
## Popular.fctr NA NA
## PubDate.last1 NA NA
## PubDate.last10 NA NA
## PubDate.last100 NA NA
## PubDate.month.fctr NA NA
## PubDate.POSIX NA NA
## PubDate.year.fctr NA NA
## PubDate.zoo NA NA
## S.has.http NA NA
## S.nchrs.log NA NA
## S.npnct02.log NA NA
## S.npnct05.log NA NA
## S.npnct10.log NA NA
## S.npnct11.log NA NA
## S.npnct14.log NA NA
## S.npnct17.log NA NA
## S.npnct18.log NA NA
## S.npnct19.log NA NA
## S.npnct20.log NA NA
## S.npnct21.log NA NA
## S.npnct23.log NA NA
## S.npnct24.log NA NA
## S.npnct25.log NA NA
## S.npnct26.log NA NA
## S.npnct27.log NA NA
## S.npnct28.log NA NA
## S.npnct29.log NA NA
## S.npnct31.log NA NA
## S.npnct32.log NA NA
## S.nwrds.log NA NA
## S.nwrds.unq.log NA NA
## S.T.articl NA NA
## S.T.compani NA NA
## S.T.fashion NA NA
## S.T.first NA NA
## S.T.intern NA NA
## S.T.newyork NA NA
## S.T.report NA NA
## S.T.week NA NA
## S.T.will NA NA
## S.T.year NA NA
## UniqueID NA NA
## WordCount NA NAsav_entity_df <- glb_entity_df
print(setdiff(names(glb_trnent_df), names(glb_entity_df)))## character(0)for (col in setdiff(names(glb_trnent_df), names(glb_entity_df)))
# Merge or cbind ?
glb_entity_df[glb_entity_df$.src == "Train", col] <- glb_trnent_df[, col]
print(setdiff(names(glb_fitent_df), names(glb_entity_df)))## character(0)print(setdiff(names(glb_OOBent_df), names(glb_entity_df)))## character(0)for (col in setdiff(names(glb_OOBent_df), names(glb_entity_df)))
# Merge or cbind ?
glb_entity_df[glb_entity_df$.lcn == "OOB", col] <- glb_OOBent_df[, col]
print(setdiff(names(glb_newent_df), names(glb_entity_df)))## character(0)if (glb_save_envir)
save(glb_feats_df, glb_entity_df,
#glb_trnent_df, glb_fitent_df, glb_OOBent_df, glb_newent_df,
glb_models_df, dsp_models_df, glb_models_lst, glb_model_type,
glb_sel_mdl, glb_sel_mdl_id,
glb_fin_mdl, glb_fin_mdl_id,
file=paste0(glb_out_pfx, "prdnew_dsk.RData"))
# tmp_replay_lst <- replay.petrisim(pn=glb_analytics_pn,
# replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
# "data.new.prediction")), flip_coord=TRUE)
# print(ggplot.petrinet(tmp_replay_lst[["pn"]]) + coord_flip())
glb_chunks_df <- myadd_chunk(glb_chunks_df, "display.session.info", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 16 predict.data.new 9 0 365.326 374.243 8.917
## 17 display.session.info 10 0 374.243 NA NANull Hypothesis (\(\sf{H_{0}}\)): mpg is not impacted by am_fctr.
The variance by am_fctr appears to be independent. #{r q1, cache=FALSE} # print(t.test(subset(cars_df, am_fctr == "automatic")$mpg, # subset(cars_df, am_fctr == "manual")$mpg, # var.equal=FALSE)$conf) # We reject the null hypothesis i.e. we have evidence to conclude that am_fctr impacts mpg (95% confidence). Manual transmission is better for miles per gallon versus automatic transmission.
## label step_major step_minor bgn end elapsed
## 6 extract.features 3 0 43.023 177.899 134.876
## 8 select.features 5 0 182.126 234.906 52.780
## 10 fit.models 7 0 236.209 275.696 39.487
## 11 fit.models 7 1 275.697 312.670 36.973
## 2 inspect.data 2 0 14.682 33.457 18.775
## 14 fit.data.training 8 0 335.359 354.080 18.721
## 12 fit.models 7 2 312.670 328.058 15.388
## 15 fit.data.training 8 1 354.080 365.325 11.246
## 16 predict.data.new 9 0 365.326 374.243 8.917
## 13 fit.models 7 3 328.059 335.358 7.299
## 4 manage.missing.data 2 2 37.516 42.965 5.450
## 7 cluster.data 4 0 177.900 182.126 4.226
## 3 cleanse.data 2 1 33.457 37.515 4.059
## 1 import.data 1 0 12.727 14.681 1.955
## 9 partition.data.training 6 0 234.906 236.208 1.303
## 5 encode.data 2 3 42.966 43.023 0.057
## duration
## 6 134.876
## 8 52.780
## 10 39.487
## 11 36.973
## 2 18.775
## 14 18.721
## 12 15.388
## 15 11.245
## 16 8.917
## 13 7.299
## 4 5.449
## 7 4.226
## 3 4.058
## 1 1.954
## 9 1.302
## 5 0.057
## [1] "Total Elapsed Time: 374.243 secs"
## label step_major step_minor bgn end elapsed
## 2 fit.models_1_glm 2 0 279.831 298.839 19.008
## 3 fit.models_1_rpart 3 0 298.839 312.663 13.824
## 1 fit.models_1_bgn 1 0 279.819 279.831 0.012
## duration
## 2 19.008
## 3 13.824
## 1 0.012
## [1] "Total Elapsed Time: 312.663 secs"
## R version 3.1.3 (2015-03-09)
## Platform: x86_64-apple-darwin13.4.0 (64-bit)
## Running under: OS X 10.10.3 (Yosemite)
##
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
##
## attached base packages:
## [1] tcltk grid parallel stats graphics grDevices utils
## [8] datasets methods base
##
## other attached packages:
## [1] rpart.plot_1.5.2 rpart_4.1-9 ROCR_1.0-7
## [4] gplots_2.16.0 caTools_1.17.1 caret_6.0-41
## [7] dynamicTreeCut_1.62 proxy_0.4-14 tm_0.6
## [10] NLP_0.1-6 mice_2.22 lattice_0.20-31
## [13] Rcpp_0.11.5 plyr_1.8.1 zoo_1.7-12
## [16] sqldf_0.4-10 RSQLite_1.0.0 DBI_0.3.1
## [19] gsubfn_0.6-6 proto_0.3-10 reshape2_1.4.1
## [22] doMC_1.3.3 iterators_1.0.7 foreach_1.4.2
## [25] doBy_4.5-13 survival_2.38-1 ggplot2_1.0.1
##
## loaded via a namespace (and not attached):
## [1] bitops_1.0-6 BradleyTerry2_1.0-6 brglm_0.5-9
## [4] car_2.0-25 chron_2.3-45 class_7.3-12
## [7] codetools_0.2-11 colorspace_1.2-6 compiler_3.1.3
## [10] digest_0.6.8 e1071_1.6-4 evaluate_0.5.5
## [13] formatR_1.1 gdata_2.13.3 gtable_0.1.2
## [16] gtools_3.4.1 htmltools_0.2.6 KernSmooth_2.23-14
## [19] knitr_1.9 labeling_0.3 lme4_1.1-7
## [22] MASS_7.3-40 Matrix_1.2-0 mgcv_1.8-6
## [25] minqa_1.2.4 munsell_0.4.2 nlme_3.1-120
## [28] nloptr_1.0.4 nnet_7.3-9 pbkrtest_0.4-2
## [31] quantreg_5.11 randomForest_4.6-10 RColorBrewer_1.1-2
## [34] rmarkdown_0.5.1 scales_0.2.4 slam_0.1-32
## [37] SparseM_1.6 splines_3.1.3 stringr_0.6.2
## [40] tools_3.1.3 yaml_2.1.13